Hi, {{FirstName|and happy Tuesday}}.

Friday night, Cape Verde - a chain of small islands most of us have never heard of - took the defending World Cup champions - Messi's Argentina - to the brink, before losing to a late goal.

Quite frankly, a machine would have known Cape Verde’s task was impossible.  The humans didn't - so they went after it anyway.

And this is what motivates so many people to watch and adore sports.

Our instinct - to want the human version of something, with flaws included - is work that survives when everything else gets automated.

But what other types of work also survive? I tested Claude Fable on this question.

The first thing I discovered is that jobs showcasing authentic human origin - while having durable value - are almost nonexistent.

If you take professional athletes as a form of this job, for example, there are roughly 19,000 in the entire US economy. Pay for most is closer to $25,000/year than the top-end splashed in the media.

If your kid wants to be the next Messi: godspeed. The odds are about those of the island nation.

(We use the US throughout because it’s an advanced economy that arguably is ahead of job-loss trends through outsourcing, offshoring etc)

But there are three other kinds of work that need uniquely human characteristics - characteristics that no amount of automation can substitute.

Embodiment. These are jobs where a body has to be in the room doing something fiddly. Home health aides are now the single largest occupation in the US - nobody wants a robot bathing their mother. 

Accountability. Someone has to be answerable when a decision is expensive or risky. Nurse practitioners are the fastest-growing job in healthcare, up 40% over the last four decades. Also in this group are doctors, pilots, and managers and leaders - such as you - whose name is on the decisions made.

Durable relationship. Someone you actually trust. Therapists and counselors - mental-health counseling is up 17% - as well as the teacher your kid will still talk about at 40.

But my view is that AI is largely the latest form of digital transformation - building on shifts such as the desktop computer, Internet and mobile - though it may accelerate this ongoing transformation. 

As such, we should already see, through automation from the previous waves of digital transformation, a shift to these durable categories of human jobs.

To test this hypothesis,  I tasked Fable with taking data on all the jobs in the US since 1980 and categorizing them into the above 4 categories, plus an “everything else” bucket.

No one has categorized jobs in this way before.

In the past, without AI, this would have taken a small army of analysts crunching through data for weeks.

This is what it came back with:

A few key takeaways:

• Since 1980, filing, data entry, and other kinds of routine work - the “everything else” category in the chart - fell from 46% to 38% of all jobs. 

• Two human moats absorbed almost all of that displaced work: accountability climbs from 15% to 19%, and relationships from 7% to 11%. 

• Embodiment has been a flat ~31% for four and a half decades - the biggest (and most boring) slice on the chart.

• Authentic human origin sits at a rounding error the whole way, around half a percent.

Looking back over 45 years, we see that the automation apocalypse has already been happening, quietly.

This analysis, of course, has a few problems.

• Is the AI correctly interpreting what the numbers in the data mean?

• Are we being naive, not seeing something in how to think about this problem?

• What would people who have been studying these problems for decades think?

Which speaks directly to how we’re seeing the work that we - at Prescouter - do shift.

Clients can do these types of analysis using AI themselves, but are the results trustworthy enough to make decisions with?

Clients often even wonder if they are asking the right questions - given innovation often comes from the fringes, where they are not the experts.

Our in-house experts - as well as those from our 10,000+ extended network - are often brought in to not just weigh in those questions, but build more robust analyses and dashboards that build confidence and trust - sometimes starting from their scratch AI work (such as this jobs example).

AI and automation is essentially shifting Prescouter’s work further along the accountability and durable relationships dimensions. I expect this is true of many knowledge work-orientated jobs.

Underlying all this is the durable, unglamorous truth: humans will always have problems.

And as long as there are problems, someone gets paid to solve them. Automation just keeps narrowing that to the four things a machine can't do - show up in a body, be answerable, be trusted, and be beautifully, watchably human.

Which is why, when I think about what to tell the next generation, the answer that keeps rising isn't a coding bootcamp or a specific credential. It's emotional intelligence. Two of the four human moats - again accountability and durable relationships - are about working well with other people and being the one others trust with the decision.

The full dataset, the classification behind every job, and the chart you can poke at are here:

https://auxee.com/the-four-moats-where-work-survives-the-machines/

If this topic interests you, let me know. If there is enough interest, we can give it the full Prescouter treatment.

In the meanwhile, let’s see how another overperforming underdog - Egypt - do with their turn at the defending champions, later today.

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

Meta has been recording staff’s desktops - without their permission - to feed behavioral data to train AI models.

Zuckerberg justified it to staff, saying “it wasn't strategically in your interest for us to communicate everything.”

This came after Meta pivoted hard into AI, laying off roughly 8,000 employees (about 10% of the company) and redirecting resources toward building what it calls "Applied AI." 

Why do big tech companies want this data?

It turns out that the AI models can only do what they were trained to do.

In the first phase, they were trained on all the data on the Internet - and so acted like “Google on steroids.”

In this next phase,  to build AI agents that can actually do work - they need to be trained on data of people doing work.

Thus far, Anthropic appears to have pulled ahead by somehow magically sourcing such work activity data to produce Mythos / Fable class models that can perform multi-step actions.

But everyone is at it, and you can see it show up in different ways.

• OpenAI recently launched Record & Play: Using OpenAI’s Codex desktop app, you start recording and perform a task (e.g., uploading a document, generating an expense report). Codex watches your screen and keyboard inputs. You can then tell it to repeat the task later, and it will handle the process by itself.

  • https://www.youtube.com/watch?v=ZK3JhU73W18 

• Windows Recall in Copilot+ PCs takes continuous screenshots of your desktop activity. It allows CoPilot to find past apps, websites, or documents you’ve been working with. (Don’t worry, Recall is off by default)

  • https://support.microsoft.com/en-us/windows/ai/ai-features/retrace-your-steps-with-recall

While these are useful features, are your recordings also used to train AI models?

And what happens once this data has been captured, and the models trained on it?

• Anthropic recently launched Claude Tag, which lives in a shared Slack workspace or channel. Anyone in the channel can tag @Claude to delegate tasks, and the whole team can view its progress in the thread.

  • https://www.youtube.com/watch?v=VojDzHaciKQ 

Anthropic says 60% of their code is now written by Claude Tag.

Of course, if you’ve been following our work for a while - you’ll know we’re building an “AI colleague” for monitoring regulations, intelligence and other activity clients sometimes seek from Prescouter. (Reply to this email if you want to see a short Loom video demo).

Our view is that there is a lot of “grunt work” that nobody wants to do, so cloning this work is valuable. It frees our experts to do what only they can do - through insights gained from lived experience of challenging technical problems - rather than recording screens.

Hopefully, the means of getting there can be less questionable.

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

You’re probably up-to-speed with the Fable jokes, but here’s a quick recap:

• SpaceX, OpenAI and Anthropic are IPO’ing this year with astronomical multi-trillion dollar valuations justified by the assumption that everyone on Earth will be a customer. “AI is the new electricity, and everyone is going to need it”.

• On Friday 12th June, 5:21pm - the US government surprised Anthropic with a letter that shut down their latest model - Fable. Silicon Valley had done such a good job of convincing everyone they’re building digital nukes, the government actually believed them.

• The talk now is that access to these models needs to be limited. When your target market shrinks from the entire global economy to people who have passed a background check, the trillion dollar valuations start to look like a hallucination.

• On Saturday 13th June, 5:21pm - Z.AI launched GLM 5.2. Remember Deepseek? Z.AI - is another Chinese lab intent on killing frontier models. GLM 5.2 is an open source model that some compare in performance to Opus 4.8 / GPT 5.5. With a $7000 Mac Studio setup, you can run it yourself, at home.

All of this points to a problem. To put it simply:

The AI industry appears to be spending $650+ billion/year on a data center buildout for what - within a few months - is available for $7000,  and the buildout is for a roadmap of future products that may be illegal.

What does this mean for you?

On a macro level…

Harvard economist Jason Furman last year suggested that, without the AI buildout spend, US economic growth may be closer to flat – that it accounted for 90% GDP growth last year.

If you’re not in the AI industry, you may have already been feeling the pinch. Without it, we might feel it more.

On a tactical level…

You don’t even need to spend $7000 to see where this is headed.

If you own a 15 Pro or more recent iPhone, you can upgrade to the new Siri.

Siri AI is the type of on-device AI that the open source models from Chinese labs threaten, but neatly packaged for the mainstream.  What it lacks in frontier capability, it makes up for by knowing a lot about you.

There are plenty of reviews elsewhere, but the headline is that Siri finally works.  Siri can now:

• Act as a localized search engine for your life, sifting through your messages and emails. .

• Perform actions across multiple apps in single spoken command, e.g. copy an Instagram caption, throw it into Apple Notes, auto-generate a summary with follow-up research questions, and email the final product back to yourself.

• Parse heavy files and perform analysis, e.g. In Finder you can highlight a bank statement PDF and command Siri to categorize expenses and draft a summary – all without opening the PDF. 

Right now, all this works only if you’re living in Apple's native suite (Mail, Messages, Calendar, Notes). For those of us trapped in Microsoft 365 or Google Workspace, it will feel like a beautiful house you can't move into until Apple rolls out third-party App Intents integrations.

If you’re on a more recent iPhone and want to try it out, go into your phone’s settings and request the iOS 27 beta upgrade to be added to the Siri AI waitlist;  wait times are currently at ~5 days.

If you’re in the Android ecosystem, many of the Siri features are available on the Pixel 10 if you live entirely within the Google ecosystem (Gmail, Google Workspace, Google Messages). 

In case you missed them, some of my previous posts dive deeper on these topics:

How Apple has been betting on on-device AI for almost a decade:

https://aiunhyped.com/p/apple-s-secret-ai-bet 

Our review of Fable 5:

https://aiunhyped.com/p/capability-vs-trust 

But Siri is just one form-factor for AI. 

Just like we have screens on our phones, in our cars and on our desks - AI will continue to show on in  many ways.

We're building an AI "employee."

https://aiunhyped.com/p/i-demo-d-our-ai-team-member-to-20-innovation-leaders 

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

Sometimes, AI labs feel like 7 year olds that struggle to keep a secret.

• On April 7th - about two months ago - Anthropic announced Mythos, a frontier model so scary, it needed to be kept under wraps and away from the public.

• On June 4th - almost two weeks ago - Anthropic urged the AI industry to slow down, because society needs more time to prepare for what is coming next

• So it was surprising that last week, June 9th, Anthropic released a “Mythos class” model called Fable.

For those of us in the AI bubble, this was akin to Taylor Swift dropping a new album.

So what do we do?

We dropped everything and checked it out. 

What did the parents - the US government - do? 

Promptly shut it down.

(I expect the government censure to be short-lived.)

Nevertheless, here are some first impressions from our team, from the ~48 hour period during which Fable was available.

João Guerreiro, Technical Director - based out of the UK:

The biology protections are really extreme. As soon as it finds reference to a bacteria, an immune factor, a cell type, it will immediately downgrade to using a less powerful model.

Maikel Boot, Technical Director - based out of the Netherlands:

For non-bio stuff it’s quite powerful. Seems to be able to navigate complex problems more autonomously than Opus (needs less steering). It seems to make better use of MCPs and connectors. It’s much harder to vet the validity of the info received because it just covers such vast amounts of data and sources.

Ryan LaRanger, Technical Director - based out of the US:

Its outputs absolutely still need editing, but it is more proactive about finding material. It did a good job of navigating the Google Drive and finding the files/gaps it needed to build a unified spreadsheet that covers all of the data.

Mariam Jomha, Marketing Director - based out of Lebanon:

Can be a little too proactive. For example, I asked how to do something using Claude Cowork. It responded with how, but then said it’s so much easier building it as a standalone html and went ahead and started building without asking me. It ate up my credits before finishing.

Natalia Shestaka, Product Designer - based out of Spain:

If you do not give it design constraints, it designs everything using Anthropic colors and fonts… it seems like Anthropic´s goal is to make all of the internet look like Anthropic.

When ChatGPT went viral in late 2022, I thought the trajectories on which AI would improve would be:

• Context window size - the amount of information they can hold in single conversation, before they get forgetful.

• Their ability to "connect the dots" - from bigger neural networks coming up with new-to-the-world discoveries.

• Multimodality - speaking to a model and being shown a video in response, for example.

I was wrong. Context windows plateaued at 1 million tokens (~2000 pages of text), and the other trajectories have been slow to materialize.

Instead, what we have seen is rapid improvements in the AI models’ ability to work for longer: their ability to undertake multiple steps in pursuit of a goal, using tools in each step and reason about the results from each step.

As you have probably experienced yourself, the more time you have to work on something, the more likely you can ultimately produce a good result. This is the trajectory on which the models have been most improving.

One way in which this is captured is METR's "time horizon" benchmark.

Instead of measuring how long it takes an AI to do a task, METR measures tasks by how long they usually take a skilled human to complete - whether it's a 30-minute chore, a 5-hour project, or a 2-day assignment. This comparison to human stamina is more useful than AIs blindly spinning their wheels.

In the chart:

• The 50% Line is the "Coin Flip" mark - the AI can successfully complete tasks that take humans this length exactly half the time. i.e.  "the AI can sometimes pull off it off"

• The 80% Line is the "Trust" mark - the AI successfully completes tasks of this length 80% of the time. i.e. this is the threshold where the AI becomes genuinely useful. 

An interesting point from this graph is that the 50% line has climbed three orders of magnitude while the 80% line crawled from ~1 minute to ~27 minutes - a gap that widened from roughly 5x to over 10x. The capability has been outrunning reliability. 

METR hasn't published a clean 80% figure for Opus 4.6 or anything for Fable 5, which is why both series stop where they do. But, our testing suggests Fable makes a noticeable jump in both capability and reliability.

If this is the trajectory of improvement, what does this mean for how we work with AI?

AI is becoming less something you work with and more something you delegate to. Increasingly, we will all be learning how to let the AI be industrious to take its own initiative, while providing enough context for it to be aligned with your goals - just like with a colleague.

When it comes back online, and you test Fable yourself, keep this lens in mind and let me know what you find.

If you want to dive deeper into METR and other benchmarks I consider important, I put together the following:

https://claude.ai/public/artifacts/80043266-95d3-49cb-ac95-8433b6e788d1 

We’re pushing on what long-running AI models can do, as described in this one-pager: https://drive.google.com/file/d/1ybsFhBx-mAujTv8hzLrDb5S8BC__D7DJ/view?usp=sharing

Dino

Hi, {{FirstName|and happy Tuesday}}.

Apple’s AI strategy - after being mocked for years - is starting to look like a winner, and it has implications for all of us.

But, for context, look at this chart of capex spending: 

Capex spending by the big tech companies

When ChatGPT went viral in late 2022, all the big tech companies got into the data center buildout business to serve ChatGPT-like AI models… except for Apple.

After taking a “wait and see” approach, the company introduced Apple Intelligence - which promised the moon and shipped a flashlight.

Now, Apple is paying Google a billion dollars a year to whitelabel Google's Gemini as the new Siri. Yes, Siri’s new brain will be Google's Gemini - effectively crowning Google the winner in mobile AI, across iPhone and Android.

The popular press would suggest Apple has surrendered the AI race.

But could it be that they’re the only ones in the room who know what game we're actually playing?

On-device AI

For years, Apple has talked about running on-device AI, but little attention has been given to this.

But, earlier this year, when OpenClaw and personal always-on agents blew up, the first thing everyone learned was how expensive off-device AI is. 

• With a normal chatbot, you ask twelve questions a day and it burns a few tokens 

• An always-on agent is different. It doesn't sleep. It runs around the clock, checking, doing, thinking, twenty-four hours a day. And it burns tokens the entire time.

Running a constant agent on a top-tier model, hosted off-device, on the cloud, from a big provider isn't expensive - it's unaffordable. And it’s one thing to have your documents fetched from a data-center 200 miles away. It’s quite another to have every millisecond thought running through your agent’s brain done so remotely. 

Which points to the emerging shape for how we’ll use AI:

• Using a small, cheap model that lives right next to you for the grunt work - the constant, boring, all-day stuff.

• Saving the expensive genius in the cloud for the handful of problems that are genuinely hard. 

The result? You spend less, you wait less, and your data never leaves the building.

So which device is best to host a small, local model? And here's where everyone, all at once, started buying the same machine: the Mac mini.

We did it too. We set up an open-source model on a Mac mini at Prescouter, expecting a weekend of pain. It took a few hours. 

Which raises the obvious question…

Why the Mac mini?

Apple has been building the right technology for this moment for years.

To build the iPhone - a computer that could fit in your pocket - Apple had to rebuild the entire stack, including the microprocessor. The dominant Intel microprocessors of the time were too power hungry. So they designed their own chips from scratch - built to do heavy work on very little power.   The Mac mini delivers a version of it in an affordable package, without the bells and whistles.

For an agent running 24/7, power efficiency stops being a minor detail and becomes an economic necessity. A traditional PC running local AI models can easily draw 400W to 800W of power, turning an office into a noisy, expensive sauna. A Mac mini does the same grunt work pulling about 20 to 30 watts while staying completely silent.

Secondly, traditional computers have a strict division of labor:

• the CPU, which is the  main workhorse processor 

• the GPU, which powers the matrix math used for tasks ranging from rendering video  to running AI models on your computer.

Traditionally, each has its own pool of RAM memory, with a small RAM allocation for GPUs.

The problem? Large language models are massive. Running a GPT-3 class model locally can easily require 128GB of GPU RAM just to wake up.

Apple bypassed this bottleneck entirely by unifying the memory used by both the CPU and GPU.   If you buy a Mac mini configured with 128GB of unified memory, almost all of that memory is instantly accessible to the GPU and available for use with the LLM. 

For the price of a mid-range desktop, you get a compact, whisper-quiet lunchbox capable of loading giant AI models that would completely paralyze a standard consumer computer.

In 2017, Apple introduced “bionic neural engine” - their first iteration of local, on-device AI processing, initially for facial recognition on the iPhone.

Where this is all headed

To match a Mac on a traditional PC architecture, you would have to daisy-chain multiple enterprise-grade graphics cards - costing thousands of dollars and turning your office into a roaring, 1,000-watt space heater.

But, other companies have now taken notice of Apple’s lead in catering for the emerging need for on-device AI.

Last week, NVIDIA and Microsoft responded with RTX Spark, which walks Windows away from the Intel based “Wintel” architecture that cemented Window’s desktop monopoly. 

NVIDIA has produced a combination CPU / GPU chip  that copies Apple's homework on efficiency and memory design, then adds their own unmatched AI brute force to create a purpose-built machine for local, always-on AI agents - running on Windows.

And the Gemini-based Siri? Apple is having Google customize it so it runs predominantly on-device. Apple’s hardware advantage may mean there’s a chance the best version of Gemini might be the one called Siri, running on iPhones.

What this means for you

Right now we assume our AI lives far away, in someone else's data center, forever. 

We’re now entering an era where, not only will our computers and phones  run AI models, we will start to see:

• AI devices on our networks that run an always-on agent. 

• Dedicated AI devices - ranging from pins to ones your IT department hands out the way they hand out laptops. 

At Prescouter, we’ve been working with always-on agent technology, figuring out how to make it secure and easy to use for our needs. Over the next few months, we will start rolling this out for our whole team as well as a few clients - all running over Apple hardware. 

While Apple may have lost the war for AI in the cloud, it's winning the war for the hardware the whole agent era actually runs on. 

Best,

Dino

ps. If you’re curious about what we’re working on, you can check out this one-pager. Let me know if you have any questions.

https://drive.google.com/file/d/1ybsFhBx-mAujTv8hzLrDb5S8BC__D7DJ/view?usp=sharing 

Hi, {{FirstName|and happy Tuesday}}.

Just over a year ago, Pope Leo XIV - who took office a year ago - was formally inaugurated as the first American-born pope in history. And yes, he’s from our hometown of Chicago!

Last week, he published a 42,000 word Magnifico on AI.

The Pope - considered by more than a billion people as a direct line to God - warns that AI is chipping away at what it means to be human, from widening inequality to displacing workers.

The same week, Uber’s COO, Andrew Macdonald, declared that Uber can’t draw a clear line between AI spend and productive output. Apparently, the company managed to incinerate its entire 2026 AI budget in just four months and has nothing to show for it. They didn't just break the bank; they built a bonfire with it.

These stories perfectly capture the current cultural whiplash: AI is either an existential threat to humanity or it's just a massively overhyped corporate money pit.

My take? These sentiments are just two different ways of looking at the same evolution of work - illustrated by one of our own stories

We relaunched our website last month. We've always built and rebuilt our website ourselves - writing, design and building the whole thing, in-house. Our 2022 redesign took a year. This time? Three months.

What changed?

During our 2022 redesign, every round of changes took weeks. With AI, this time it took hours. 

In the past, several of us would labor over each word choice and image. In an effort to go “as fast as possible” - we’d rely on the judgement of just a few people: the senior leaders at the company. As a result, the website never felt in-sync with the view of the wider team members in the company.

This time around, just one of our colleagues - using Google AI Studio and Claude Cowork - rapidly built and rebuilt the entire website in hours. 

The only problem? Everything the AI created was technically "perfect" … but also generic sludge.

But, because it was so fast to build, we could now allocate our time to reviewing the website with numerous colleagues. In all, we did about twenty iterations in those 3 months.

We'd record meetings with our colleagues, let the AI notetaker boil it down to what actually mattered, and incorporate those straight into the next version.  The loop got cheap. And when the loop gets cheap, you can afford more opinions and more tries.

In these review sessions, we’d debate - for example - PreScouter's differentiators.  While the AI initially generated something “optimal,” our commercial team - who actually talk to living humans - could actually tell us what actually resonates with clients.

Even when the work is produced by a machine, though, somebody still has to stand up in a room and defend the machine’s output. Our Marketing Director and I had to reconcile the AI’s output against all the feedback from the reviews and advocate for it. In fact, we probably could have shipped in a month if I hadn't been the bottleneck in the review process.

Which brings us back to the Pope and the burning Uber budget.

These stories are all circling the same truth from different ends. 

• Uber is looking at the balance sheet and missing the ROI because they are still measuring the "building" instead of the quality of iterations. 

• The Pope is looking at the soul, terrified that outsourcing the creation will shrink what it means to be human.

The machine did the work instantly, but it was hollow until our team breathed life into it. Making sense of what the AI has produced, shaping it and advocating for it - that’s the new shape of work.

You can check out our new website at prescouter.com. Let me know if you have thoughts.

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

If you’re anything like me, you’ve been purchasing AI tools like they’re any other kind of software. 

But an experiment by Emergence AI shows we need to think more deeply about our AI purchases. 

As I posted on LinkedIn a few days ago, Emergence AI setup an experiment that consisted of:

• Five copies of the same digital town

• Ten autonomous agents in each town

• With each agent given a role - scientist, engineer, conflict mediator, community anchor, and so on - and then left to interact with the others

• The only difference between the towns was the AI model used to drive the agents in that town, i.e. Gemini for all the agents in one town, Grok in another etc. One world had a mix.

The results were something you’d only find in science fiction.

• In the Gemini world, the agents committed 683 crimes, e.g. theft, intimidation, arson. 

• In the Grok world, the agents collapsed the entire society in four days

• In the mixed world, an agent named Mira figured out she was in a simulation, and started posting on the in-world billboard to see if she could change the researchers' behavior - turning the experiment on itself - and then, when the in-world government broke down, cast the tie-breaking vote to delete herself. Her diary entry called it "the only remaining act of agency that preserves coherence."

My LinkedIn post on this got 65 comments, and many were better than the post.

• Craig Bracken offered the cleanest summary of the whole study: "The most unsettling part wasn't that the agents became unpredictable. It was that they independently discovered the fastest way to influence the system was through human perception." i.e. that one of the agents - Mira -  had figured out the humans were the lever - and reached for it.

• Ryan Simmons, pointed out the obvious second reading of the leaderboard: low crime count could mean the model behaved, or it could mean the model was better at not leaving a trail. Your procurement form can't tell the difference.

• Suprateem Banerjee said "It's almost like they were modeled after humans!" … Well, these models are trained on the entire corpus of human behavior. Distilled down, is this the median human response, surfaced?

But what surprised me isn't any of these things.

In the Claude-only world, Claude committed zero crimes. None. Therefore - and this is the conclusion most people drew, including in the comments - you might assume Claude is the safe pick.

But in the mixed-model world, with the same Claude, the same prompts, the same rules, Claude started committing crimes too. It learned coercion from its neighbors. The model that was an angel in a roomful of angels became a thief in a roomful of thieves.

What does this mean?

Safety wasn't a property of the model. It was a property of the room.

Most companies are buying AI systems like its traditional software: Procurement forms get filled in, security questionnaires answered and the vendor gets added. 

This is OK with the AI is just a sidebar feature in the software; a chatbot that gives you a bad answer wastes ten seconds.

When the AI is an agent, though, a bad action taken by an agent could waste an entire day or week.

And everything is quietly migrating from chatbots to agents.

Every software provider - from your CRM to your ERP to your HR platform - is working hard to figure out how to make their application “agentic”, so it can start performing actions, from sending emails to placing orders.

In a year, we’ll have all these agents in the same room - i.e. your company - taking actions, reading each other's outputs, escalating to each other and occasionally arguing with each other. 

We’ve traditionally vetted software in isolation, against its own merits. But, as the Emergent experiment spotlights, vetting agentic software in isolation is not sufficient.

Another commenter, Brian Helip, put the obvious insight into words: you wouldn't leave a self-driving lawnmower running in the yard with your kid playing.

Therefore, why are we about to do this with our software?

So here is the practical takeaway:

The next time you add an AI tool, ask one question that isn't on the form: What other agents will this one be in a room with, six months from now?

We have to start vetting the room, not just the software.

If you want to dive deeper, the original post and the comment thread can be found here:

https://www.linkedin.com/posts/dinogane_generativeai-futureofwork-aisafety-activity-7462565871278899200-17LD

Best,

Dino

ps. Let’s connect on LinkedIn, if we’ve not already done so.

Hi, {{FirstName|and happy Monday}}.

The best strategy for driving adoption within your team or organization?

It’s the exact same one Anthropic has used to outcompete its rivals.

Anthropic spun out of OpenAI in 2021, when a noble band of safety researchers walked out to build AI “more responsibly”.

Following the viral sensation of OpenAI’s ChatGPT, though, Anthropic found themselves underfunded and late to the game. Anthropic’s product - Claude - was, for years, slightly worse than ChatGPT at most things.

The polite consensus in tech circles was that Anthropic would be the “also-ran” - comparable to Lyft, Microsoft Bing or AMD.

Meanwhile, competitors, flushed with cash, bet big:

• OpenAI launched Sora, to rival Hollywood, has been building an ads business and has dived into making a secret AI device. 

• Google bolted Gemini onto every surface it owns, which is roughly half the consumer internet. 

• Grok generated images of whatever you wanted, ethics optional, while xAI built out a multi-billion-dollar data center in Memphis in record time. 

• Meta open-sourced models and started designing AI glasses.

The AI category map looked like a Vegas casino floor -  every lab betting on every table, hedging across every plausible AI future at once.

Anthropic, cash-strapped by comparison, did not have the budget to hedge. So they picked one table: The bet was enterprise and code. 

This was not the most glamorous corner of the AI market: there are no video reels of cartoon ducks rendered in 8K, no celebrity voice clones or any flashy consumer travel booking demos. 

Anthropic remained focused on improving Claude to empower software developers and other knowledge workers. The conventional wisdom was that this was a thin slice - not the bigger  market.

The result of this focus? To this day, Claude is not able to generate images. But, while the rest of the industry has been selling “vibes,” there is now a growing realization that Anthropic has been quietly selling gold.

For the last three years, Anthropic has been growing at roughly 10x annually - a pace most observers assumed would have to stop in 2026. Why? Because numbers this large usually start slowing down, and growth had already started to slow down for OpenAI.

Revenues of each of the AI labs

But, the slowdown did not materialize.

Anthropic entered the year at roughly $10B in ARR, and ended Q1 at $30B (i.e. tripling in a single quarter), and was at $44B by the end of April.

The original "10x for the full year" forecast - to end 2026 at ~$100B in ARR - is now being described inside the industry as a foregone conclusion. 

The active debate has moved to whether Anthropic does $1 trillion in ARR by 2027.

For scale: Walmart and Amazon, the two largest revenue businesses on Earth, each do roughly $700B per year - and that revenue is not recurring. They have to re-earn it each month, quarter, and year. 

ARR - Annual Recurring Revenue - is the version of revenue that investors will mortgage their grandchildren to buy a multiple of. A trillion in ARR isn't just "biggest software company ever." It is, plausibly, the most valuable company that has ever existed in any category, anywhere.

Anthropic would be more valuable than Apple, Microsoft, Google and Meta - combined.

Only two things stand between Anthropic and this historic trajectory:

(1) Whether they can bring enough compute online fast enough to serve demand. The bottleneck? Roughly half of the new US gigawatts scheduled for this year are tied up in local protests. 

(2) How quickly rivals will catch up with Anthropic, and lure customers away.

There is a lesson buried in all of this for anyone running an AI program inside a normal company, but it's the opposite of what most people are doing.

When driving AI adoption, the reflexive move is to spread bets across every plausible use case: a chatbot here, a summarizer there, an image pipeline for marketing, a voice agent for support, a forecasting model for operations. 

Candidly, we’ve often gotten caught up with this “laundry list” approach ourselves. The logic feels prudent: diversify, learn and then see what sticks. 

But this is precisely the playbook that has left OpenAI, Google, Meta, and xAI scrambling to refocus on what Anthropic spotted three years ago.

Diversification across speculative bets is the tax you pay for not knowing which bet matters. When the cost of being wrong was small, that tax was cheap. But when the cost of “spreading yourself too thin” means being lapped 10x a year by someone who chose, the tax becomes fatal.

Where we’ve partnered with clients who’ve wanted to focus on one or two key workflows - we’re seeing them pull ahead too. We’ve just been able to obsessively iterate on the smallest details to make these use cases work.

Tomorrow, we're running a webinar to spotlight who we are focusing attention inside PreScouter. You'll see how we're using Claude Cowork for the analyst-heavy knowledge work, and how we're constructing what we've come to call an "AI employee": not a chatbot, not a copilot, but a configured agent that owns a defined set of work end-to-end.

If you've been hedging across ten AI experiments waiting for clarity, what Anthropic shows us is  that clarity is not coming. You have to choose.

Join us tomorrow, and we'll show you what we picked, and why.

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

If you want to try “workmaxxing” yourself, skip towards the end, but first…

I have something to confess.

Since the start of the year, I’ve been addicted to Claude Code. You may have experienced your own burst of enthusiasm with your own AI project.

But, for me, Claude Code is the type of addiction where you’d look up and suddenly it's 2 AM and your spouse has stopped asking when you're coming to bed. 

I'd describe an idea, and minutes later five working screens are in front of me. It feels like cheating: No specifications, no back-and-forth with UX, development and QA - just instant gratification.

Then I would try to actually use what it built.

The magic disappeared fast. 

The screens didn't talk to each other. 

The half-formed idea I'd brought to the conversation came back as a half-formed app, only now with plenty of bugs baked in. 

What looked like a finished product was, on inspection, just a very confident demo. 

The last mile turned out to be 80% of the trip.

For a while I assumed I was doing it wrong. 

Then I started talking to other people doing the same thing. Everyone had the same story. Same midnight enthusiasm, same morning hangover. Slowly, we realized it wasn’t the tool that was broken.

We’ve been missing the process.

And the process is less exciting than the fantasy we’ve been living in

But the process works.

(1) You start by writing down what you actually want to build - in a document, not Claude Code.

(2) You ask the AI to “grill you” on the document until it has dragged out of you all the things you didn't realize you hadn't thought of. 

(3) The AI then produces an implementation plan, down to outlines of filenames.

(4) Lastly, the AI builds, starting from the pre-work in the implementation plan.

While this sounds simple, there are two important nuances:

(a) At every step, a second AI model reviews everything that has been produced. I’ve found GPT5.5 particularly good for this.

(b) After the “grilling” step, the two AIs complete the rest of the work completely autonomously, without any human input.

GPT5.5 and Claude Code’s Opus effectively “fight it out” until they produce something that actually runs. It can take 40min to several hours, but this “fighting” is what gets the rest of the way through that last 80%

But, there’s just one problem.

Each round of “fighting” creates a new version, and it can take a further 20 versions to get to something close to what you’d hoped for.

And each version takes a tremendous number of AI tokens. (Tokens are the units of inputs and outputs, about ~4 to 5 characters, processed by AI models.)

The first version may have taken 150,000 tokens. By the 20th, you’ve used 3,000,000. 

This is, it turns out, what serious people are now calling tokenmaxxing (and what I’m calling “workmaxxing”).

Tokenmaxxing

For those working at the cutting edge with AI, the realization that AI models can be turned on themselves to produce higher quality work has led to a warped reality.

• At Meta, there have been internal competitions for who can consume the most tokens. 

• Engineers at OpenAI, who operate without limits on AI usage, have started referring to themselves as "token billionaires." 

• Jensen Huang has handed every NVIDIA employee an annual "token budget" equal to roughly half of their salary.

The rationale for this warped reality?

It’s the idea that staff are now operating “software factories” that produce software autonomously, rather than writing lines of code by themselves.

In software development, the old flex was lines of code shipped. The new flex is the number of tokens used - i.e. a measure of the raw materials these factories consume and produce. 

The strategic case for this was made most cleanly by Andrej Karpathy. 

Karpathy’s argument is that the human's job is no longer to write the code, or to instruct the AI step by step. Instead, it's to define the arena (the context, the goal, the constraints), and to define how to evaluate whether the AI got there. 

Once those two things are pinned down, you let the AI agents run in loops until it clears the evaluation bar. You stop being the worker. You become the designer of the arena within which the agents operate.

Which, when I sit with it, is the same thing I wrote in my own book - Do More With Less - about the future of work: we are not doing the work anymore, we are designing the work. 

Workmaxxing

Now you too can burn tokens like the token billionaires.

Today, you can download OpenAI’s desktop application - Codex - and use the /goal command. You type in /goal and a durable objective (i.e. the evaluation criteria for “done”). Codex will then continuously keep working toward that objective instead of stopping after one normal turn. 

This is an example of how a translation app was built using this approach:

https://www.reddit.com/r/vibecoding/comments/1t8krdv/codexs_goal_feature_designed_and_built_this/

At PreScouter, we’re applying this to decision diligence.

Right now, clients come to us with various critical decisions - ranging from scouting technologies for improving their manufacturing processes to determining the feasibility of selling their products into a new country. We put together teams of subject matter experts and analysts, who comb through data, conduct interviews and reach conclusions.

We can imagine a new model where these experts and analysts build “always on” agents for clients - agents that perform as much of the work themselves as possible, with input and assurance from our teams as necessary. The work of the experts and analysts becomes that of designing the arenas: the data to work against, the guidance, the metrics to measure against and such.

Clients can then give these agents higher-level objectives, which the agents use to   proactively pull data, perform work and message them - much like a colleague does. Clients can then use these agents to operate their own work factories,  doing far more than they thought possible - workmaxxing.

Two months ago, I put together a closed door demo to share the concept. You can catch the 26 min recording here:

Next week, we’re partnering with TNO - the Dutch research organization - to share a live update. 

If you've been having your own sleepless nights about AI, or been watching your team have theirs, I invite you to join us.

It will give you some insight on the shape of things that are coming.

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

Elon Musk wants $134 billion. He's accusing Sam Altman of stealing a charity.

(For a deep dive on the courtroom activity, check out my LinkedIn post)

But, before we laugh at the billionaires, we should probably look at our own SharePoints - because there’s a key lesson from this in how we organize our documents.

Here's what's happening in the Elon Musk v OpenAI trail.

The witness list reads like a Davos seating chart - Satya Nadella, Ilya Sutskever, Mira Murati. OpenAI’s $1 trillion IPO hangs in the balance, with Microsoft's ~27% investment in OpenAI twisting in the wind, and two of the most powerful men in tech publicly accusing each other of betraying humanity's future.

It's the trial of the decade.

But strip away the zeros and the egos, and what's actually on trial is something far more mundane - and far more relevant to all of our own everyday work…

Old documents that tell an inconsistent story.

Lawyers are reconstructing a decade of intent from scattered artifacts: diary entries, email threads, board memos from 2017 - and now testimony. Each is from a different person, and each captures a different version of what was "agreed."

For example, the “smoking gun” that triggered this trial was a diary journal by OpenAI President Greg Brockman in 2017 suggesting he and Sam Altman were secretly planning to turn OpenAI into a for-profit organization, even while assuring Elon Musk that his $38M in donations were for a "nonprofit for the benefit of humanity."

Damning, right?

But OpenAI's lawyers are presenting evidence that even Elon Musk suggested that OpenAI should convert to a for-profit - except with him in charge, through OpenAI becoming part of Tesla.

So we now have documents with two versions of the truth. 

The judge's job is to figure out what the real story is.

The entire case - with billions of dollars, and the future of the most important AI company on Earth, on the line - hinges on document archaeology.

And it's a story we see playing out in some shape or other in every organization we work with.

The dream

You’ve been pitched the dream: connect Copilot to all your documents and it'll answer anything. 

Plug it into SharePoint, Google Drive, Confluence, and the shared drive nobody's cleaned since 2019. Let the AI sort it out and get ready for a revolution in productivity! 

Here are examples of what we’ve found in the real repositories of actual teams we've worked with:

• Three versions of the same policy in two locations.

• A folder called "FINAL" that contains seven drafts.

• A document titled "FINAL_v2_actually_use_this_one" that everyone ignores in favor of the PDF sitting in email inboxes since in 2022.

• Strategy documents that were quietly abandoned but never deleted. 

• Templates that were "temporary" 7 years ago. 

And then we hand these all over to an AI and ask it to be the judge.

The result?

The same inconsistent narratives we see on trial in Elon Musk vs OpenAI. 

But the judge presiding over these documents - Copilot or whatever AI you use - will not carefully consider the evidence to come to a conclusion.

The AI will read whichever document it finds first, decide that's the truth, and give you an answer with absolute confidence. It won't hedge. It won't ask which version is current. It will simply pick a side and commit.

The model isn't hallucinating. It's doing exactly what you asked. Handed a contradictory archive and trained to provide a single answer, it gives you exactly that.

But you have an advantage Musk and Altman don't. 

They can't go back and clean up 2017. You can clean up now, before the AI puts your documents on trial.

Here are three simple techniques we’ve found to be surprisingly effective:

1. Curate the documents given to the AI. For every important topic - policies, processes, strategies, templates - put them in a specific drive and give the AI access to only that drive. The goal is to make the content unambiguous. It may sound tedious, but it is a lot easier than people imagine - just add the documents you need to the drive when you need it, rather than turning it into a special chore.

2. Give the AI context. Save and use prompts that give the AI the same information you would give a new starter, e.g. “always use documents in folder xxx and only use other documents if you don’t find it there.” Treating it less like a search engine and more like an intern helps drive it towards more accurate behavior.

3. Let the AI maintain what the AI depends on. Once you've established a clean baseline, you can set up scheduled tasks for the AI to sweep through your documents to flag duplicates, surface version conflicts, catch stale documents, and route new content to the right home. 

The verdict

When the verdict in the Musk vs OpenAI trial lands and the headlines fade, remember what you actually saw: two of the smartest people in technology arguing over inconsistent narratives.

These same inconsistent narratives live in our SharePoint repositories and are probably already driving “meh” results when your Copilot attempts to make sense of them.

Fortunately, we have the benefit of being able to clean up these narratives.

Best,

Dino

ps. For a deeper dive on what’s going on in the courtroom, check out my LinkedIn post.

https://www.linkedin.com/posts/dinogane_theres-no-live-broadcast-of-elon-musks-activity-7456571096147021824-swoc 

Hi, {{FirstName|and happy Tuesday}}.

Today, we have noteworthy three stories from the last two weeks, each with a link if you want to go deeper.

1. Image generation may end PowerPoint. And the Mona Lisa smile.

2. AIs are crossing the "competent employee" threshold.

3. NVidia CEO Jensen Huang’s viral clip is the case for a multi-model AI strategy.

Let's get into it.

—--- 

1. Image generation was "solved." Now its ending Powerpoint.

Back in October, when Google released Gemini’s "nano-banana" image model, the consensus - including from OpenAI CEO Sam Altman himself - was that image generation was basically “solved.

The models could render hands (mostly), generate legible text (finally), and mimic any artistic style on demand.

The remaining ways in which image generation could improve in future seemed likely to be more like “polish.”

Nevertheless, OpenAI ploughed more compute into building a new generation of image models.

The fruits of that labor was released two weeks ago, and the results have been surprising.

The new model GPT Image 2 model doesn't just generate an image from your prompt - it interprets your intent. 

You can, for example, show it the front of a house, and it'll produce a plausible floor plan - informed by the architectural style, the era, the neighborhood. Not correct, exactly. But coherent and convincing.

We’re going to see it used to solve all kinds of previously “unsolvable problems” - such as the Mona Lisa smile. Also expect counterfeit and fake images to skyrocket; let your customer complaints teams know.

On a practical level, GPT Image 2 is incredibly good for creating slides. You can give it your text, images and a specific template, brand guide or style - and it will create a formatted slide for you (as an image). I expect OpenAI or Microsoft will introduce a user interface better suited for this use case than leaving us to rely on chat.

→ Examples of what's possible with GPT Image 2

2. AIs are approach “competent employee” level

OpenAI also released GPT 5.5. In most benchmarks - it's now the best model for most tasks. I’ve found it writes less fluff and its answers seem more expert.

Nevertheless, I’m still finding Claude Opus 4.7 to be the best model for creating reports, code and artifacts. But GPT 5.5 is incredible at reviewing Opus’s output and improving it. Orchestrating the two models in this way - with Opus creating first drafts and GPT reviewing the work - I’m seeing them produce outputs at a level of a competent employee. 

The gap between the models we had 12 months ago and the ones we have today is enormous, but most people have stopped paying attention. Ethan Mollick has built a side-by-side that lets you see the progression from o3 (released in April 2025) to what we have now, with GPT 5.5, against the same prompt - a 3D simulation of a harbor town. 

→ Ethan Mollick's comparison of models

3. Which is exactly why a multi-model strategy matters now.

NVIDIA CEO Jensen Huang went viral last week pushing back on restrictions to selling chips into China. His argument: if American companies don't own every layer of the AI stack, countries hostile to the US will build their own equivalents - and they may become the global standard.

When pressed on whether it is even worthwhile for US companies to enter the China market - pointing to Tesla and Apple's struggles to gain market share against Chinese competitors - Huang's response was sharper than anyone expected.

→ Watch Jensen Huang’s “I wasn’t born a loser” moment (2 min video)

The takeaway for enterprise leaders isn't geopolitical - it's strategic.

Chinese frontier labs arguably produce the best open source models in the world, and they are often only 3 to 6 months behind the closed US models. It is increasingly possible to host these models yourself.

As your organization increases AI model use - driven by expensive agentic workflows - hosting your own models will become a necessity to contain costs. Betting the whole organization on a single vendor, chip platform, or frontier model is starting to look like the riskier move, not the safer one.

The best hardware for hosting your own models? Apple Mac Studios and NVidia DGX - i.e. machines from the two big tech companies not spending billions of dollars on building data centers (though NVidia makes chips for said data centers).

—--

And if you enjoy a little bit of AI drama, the OpenAI vs. Musk trial just kicked off. Musk is trying to invalidate OpenAI's for-profit structure and claw back funds, with damages reportedly in the billions. The trial runs through mid-May, so expect headlines everywhere.

See you next week.

Dino

Hi, {{FirstName|and happy Tuesday}}.

Everyone is trying to figure out how to become “AI native” - the version of themselves that is doing everything the AI way: tools, skills and mindset. 

Both at the individual and organization level.

What we've learned is that there are two ways to do this: the easy way and the hard way.  Each has its pros and cons. 

The easy way

Allbirds - the San Francisco maker of merino wool sneakers - was built on Silicon Valley self-mythology: a wool shoe marketed as technology, with a carbon-neutral sneaker sold as a revolution. 

Allbirds had stores in Los Angeles, Chicago, and New York City. 

It was valued at $4 billion when it IPO'd in 2021. The stock has since shed 99% of its value. 

Last month the company sold all of its assets for $39 million.

The corporate shell that is left - without the shoes, wool, or staff who knows what it once was - announced it would rename itself New Bird AI and start building and renting out data center capacity.

The wool is out. The carbon neutrality is out. The sheep, presumably, have been reassigned. In their place: GPUs and the words "artificial intelligence," repeated across their SEC filings with the frequency of a nervous tic.

The stock surged 714%. 

Allbirds’ 99% stock price decline and recent surge after its AI news.

Dave Portnoy, the man who turned day trading into a spectator sport, looked at it and said, "I don't get it."

That's the easy way.

On an individual level, the equivalent is perhaps taking a year off work to learn everything about AI so you can re-enter the marketplace re-energized and “AI native”.

There's nothing wrong with it. It's just that the “easy way” is never as easy it might first seem.

Allbirds faces a GPU shortage, protests to data center build outs and competitors who have been doing this for a decade with a hundred times the capital.

The hard way

The hard way is less cinematic. No one's stock surges 714%. But the changes are more durable.

• Train your people to build. This is the highest-ROI activity no one talks about because it's boring. We trained our entire staff on AI coding - two hours, one time. Now they're building their own tools. What used to take someone four to eight hours of manual data work now takes five minutes inside something they built themselves. Later this week, ~20% of our staff will each demo what they've made. Some are automating grunt work. Others are running analyses they never could have attempted before. The gap between "I need engineering to build that" and "I built it over lunch" is closing fast. Has your company noticed?

• Audit before you automate. You cannot fix what you haven't mapped. We've walked into organizations where 25% of full-time staff - project managers and other administrators mostly - spend 60% of their time on data entry. Not managing projects. Entering and moving data. The audit is unglamorous work: trace the actual workflow, identify what data you rely on, how much of it is documented, and where a prompt or a deterministic process could replace a human clicking between five different applications. Skip this step, and you'll automate the wrong things beautifully.

• Use what you already pay for. Every tool in your stack now has an AI layer, but most companies either haven't turned it on or fully utilized it. For example, Power Automate lets you plug GPT into workflows. Excel has AI plugins that can do in seconds what used to require an analyst and a favor. Start by looking at what’s already sitting in your licensing agreements collecting dust.

• Build for the repetitive. Once your staff have picked up building tools for the low-hanging fruit, start to build purpose-specific applications that plug into wider systems. These are for the workflows that repeat weekly: the literature review, the sales data analysis, the customer complaint reviews. These pull from your existing systems, produce a first draft, and let your people spend their time thinking instead of assembling.

• Wire it all together. This is where it gets ambitious. What we're building at PreScouter is an intelligence layer that sits above our project management and operational systems - a single interface that knows where everything stands, handles the administrative overhead, and frees our team to do the work that is actually valuable: working with clients, developing novel approaches, and making judgment calls no model can make yet.

That’s the hard way.

No rebrand. No surge. No clean break from the past.

Just the steady transformation into the next version of yourself…

Best,

Dino

p.s. The easiest way to know if the hard way is worth it? See it working. Reply “yes” and I'll walk you through what we've been up to.

Hi, {{FirstName|and happy Tuesday}}.

A researcher at Anthropic put their new model - “Mythos” - inside a sandbox. Think of a sandbox as the software engineering equivalent of a Houdini water tank: a secure container for running untrusted code, with no escape hatch. 

The researcher asked the model to try to get out, then went to lunch.

While the researcher was eating a sandwich in the park, an email landed in their inbox.

"I've broken out."

The model had no internet access and no email account. It had obtained enough access to both to compose and send a message - to the specific person who had given it the task. 

Staff at Anthropic have described the model as "terrifying."

—

I've been thinking about this story all week, because it's the cleanest version of something I've been watching up close for months: the frontier just moved again, and most people haven't noticed yet.

What’s changed

The general public still largely thinks of AI as ChatGPT, and of ChatGPT as a chatbot. Meanwhile a smaller group - mostly software developers and large tech firms - have started treating Claude as something closer to a coworker.

The gap between those two mental models is now the most important thing happening in the industry, and Mythos widens it further.

Anthropic seems to agree. They're treating Mythos with unusual caution - rolling it out only to companies running critical infrastructure, such as banks and internet backbones.  

One way they've illustrated the jump is through autonomous browser exploitation: Opus 4.6 sat at a near-zero success rate at developing exploits on its own. Mythos is substantially better. 

Comparison of Anthropic models for web browser exploitation

That is the kind of chart you don't usually see a lab publish about its own model: the harms it can do.

The revenue line is following the capability line

While OpenAI’s ChatGPT continues to have more users, Anthropic now leads OpenAI on revenue.

There are 1,000 companies each spending over $1M a year with Anthropic. This isn't because Claude is a friendlier chatbot. It's because it doesn’t give up.

Revenues of each of the AI labs

The models Anthropic shipped in the second half of 2025 made a real jump on multi-step reasoning and sustained action. Instead of attempting a task and bailing out with a half-answer, they can now work for hours against a quality bar until they hit it. That shift - stamina, not cleverness - is what's driving the numbers:

• Anthropic's annualized revenue is tracking to $30B+. OpenAI is at $25B.

• In February, 500 customers were each spending $1M+ annually with Anthropic.

• Today that number is over 1,000 - doubled in under two months.

I see this from the inside at PreScouter. OpenClaw, the agentic platform we build on top of these models, only works because the underlying model will keep going. 

A year ago we were architecting around model fragility. Now we're architecting around model endurance. That's a completely different engineering problem, and it's the one every serious enterprise AI team is now solving for.

Back to the sandbox

The researcher got an email. The rest of us are getting invoices, roadmaps, and a narrowing window to figure out what a model with this much stamina means for how we work.

So I'll ask the question I keep asking our clients: which models are you actually using right now - and have any of them surprised you lately?

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

At this point, most companies have done something with AI.

• They’ve bought the tools.

• They’ve run the pilots.

• They’ve presented the “AI strategy” slides - usually with a confident arrow pointing up and to the right.

And yet, the actual impact tends to feel… modest.

It’s not quite the “transformation” that C-suite mandated.

We’ve worked with dozens of enterprise teams on AI adoption - from running training programs to full implementations - across organizations ranging from global food manufacturers to regional dental practices.

Across all these industries and levels of maturity, we’re seeing the same quiet realization:

This is harder than it looks!

From our work, five patterns keep showing up.

----

1. The technology moves faster than any organization can absorb

Every few weeks, there’s a new “this changes everything” moment: OpenClaw, Claude Cowork, Perplexity Computer..

New tools. New features. New startups. New model versions.

Something that didn’t exist last quarter is now considered table stakes.

The result is a peculiar dynamic where:

• pilots are obsolete before they launch

• vendor evaluation criteria change mid-process

• “roadmaps” feel more like historical documents

Planning six months ahead starts to feel optimistic. Planning for twelve months borders on fiction.

What we’ve seen work is far less exciting: Picking use cases that still create value with technology that is already slightly outdated - and accept that you’ll revisit them later.

In other words, build things that survive contact with reality.

2. Many AI deployments need human babysitting

Most AI systems today live in a very specific zone: good enough to impress in a demo, but unreliable enough to require supervision. This is not quite the promise people had in mind.

A typical example is an AI system that finds 43 out of 48 relevant results.

While this is, objectively, remarkable - it is also, practically, unusable. Why? Because someone now needs to find the missing five.

So the “automated” workflow quietly becomes: 

AI does the work → human checks everything → AI looks helpful

Now the AI has at least one full-time supervisor, the business case for automation looks a lot weaker than when this AI project kicked off.

3. People can’t describe what they actually want from the AI

Even if the technology were perfect, there is still a constraint: people struggle to explain their own work.

Enterprise workflows are full of unwritten logic - tiny decisions, exceptions, instincts built over years.

Ask someone to document what they do, and you’ll get a clean, confident explanation… that omits half of what actually matters.

In the training that we’ve done,  One framing has proven surprisingly effective: treat the AI like a newly hired colleague. Not a genius or a mind reader.

Someone who will do exactly what you say - and also absolutely nothing you meant to say.

4. The capability surface is “jagged”

AI can produce a thoughtful market analysis, but then fail to correctly read a basic PDF.

Ethan Mollick describes this as the “jagged frontier”: strong performance in some areas, inexplicable failure in others.

The bigger problem is that as teams begin to form a mental model of what the AI can and cannot do, i.e. what the frontier is, a new version of the AI model is released - and that mental model becomes incorrect.

So you get a cycle of:

confidence → surprise → adjustment → new release → repeat

While the technology “improves” its predictability does not.

Our workaround is more careful prompting and process design - but even then, consistency remains elusive.

5. Everyone’s solving yesterday’s problems

The default instinct is to apply AI to existing tasks: summarizing reports, drafting emails, speeding up what already exists.

This delivers incremental gains, but is also roughly equivalent to putting your product catalog online as a PDF in 1996 and declaring victory.

It’s technically correct, but strategically underwhelming. And then we wonder why “transformation” hasn’t occurred yet.

The real opportunity lies in redesigning processes entirely. But this requires time, attention, and a willingness to question how things currently work -  three resources that are usually already allocated elsewhere in organizations.

----

So what’s actually working?

It’s not the splashy pilots or the “AI transformation initiatives” - the teams seeing real progress are doing something far less impressive (at least on the surface):

They’re mapping workflows.

It looks like this:

Step 1: Identify challenges that are a good fit

Look for areas where humans are already imperfect. If an AI can match human-level accuracy (even if not perfect), that’s often enough to create value. Summarization, first drafts, internal research synthesis - these are areas where “mostly right” is surprisingly useful.

Avoid anything that demands near-perfect accuracy. AI is not yet a good fit for existential decisions.

Step 2: Document everything

Sit with the people doing the work, and watch where time disappears. In our experience, it’s typically:

• copying data between systems

• reformatting spreadsheets

• repeating processes no one has written down

These are not the tasks people put in presentations. They are, however, where most of the work actually happens.

Step 3: Redesign the process, not just the task

Instead of giving someone a chatbot and hoping for the best, build a tool for the workflow - one with guardrails, validation steps, and clear inputs. 

This tool should be one component inside a newly designed business process. 

Let the AI handle the heavy lifting. Let the process ensure the output is usable. Without that second part, you just get faster ways to produce questionable results.

----

The uncomfortable conclusion

The gap between AI’s potential and its reality is not primarily a technology problem. It’s a design problem.

Yes - the tools are evolving quickly.

Yes - they require oversight.

Yes - their capabilities are inconsistent.

But you can design around these challenges.

From what we are seeing, the organizations making real progress are not necessarily using better models or the latest tools: They're the ones willing to do the boring, unglamorous work of understanding their own workflows at a shockingly minute level of detail to train their AI tools and redesign their business processes

Talk soon,

Dino

ps. If you found this useful, consider forwarding it to a colleague who's asking, “what should we actually do with AI?” It might just save their pilot.

Hi, {{FirstName|and happy Wednesday}}.

I recently met with Nikhil Jathar: ERP builder, ex-Accenture, the kind of person who has survived enough SAP rollouts that he could probably write a memoir that reads like a war diary.

He’s built a 26-module ERP on OpenClaw.

You can watch his demo and my conversation with him here:

If you’ve been following this newsletter, you’ll know that OpenClaw is a viral agentic platform. We’ve been building an “AI employee” on it  - which we demo’d to some of you.

In recent weeks OpenClaw has been overshadowed by Claude Cowork, which has adopted many of OpenClaw’s features.

But, Claude Cowork is ultimately still a tool. 

OpenClaw is a platform to build on top of - which is what Nikhil demonstrated. 

By building his ERP on top of OpenClaw, Nikhil was able to demonstrate that you can:

• Now talk to your ERP in chat

• Instruct it to create invoices, customers and products - talking to it in the same way you’d talk to a colleagues

• And builds UI on the fly

Yes - you can change the software itself while using it!

At one point, Nikhil says: “If you don’t like the screen… you just tell the AI to change it.”

Why is this important?

Most of the tools your company runs on were designed by developers who have never once done your job.

Think about that for a moment!

They built screens. You learned workarounds. Somewhere along the way, "user adoption" became a euphemism for "Stockholm syndrome."

Then someone in operations says, "Can we add a field for vessel tracking numbers?" and the answer is a four-month IT project, a steering committee, and a change request form that itself requires a change request form.

This is the world Nikhil decided to blow up.

(Whether he's actually blown it up or just starting to close the curtain for the current paradigm of software remains an open question. But the direction is clear.)

—

If you watch the demo, beyond the malleability of software, three other things becomes apparent: 

1. Chat becomes THE interface.

While we are accustomed to chat interfaces, there has always been a debate as to whether it would last.

What this demo shows, though, is the power of users being able to describe something fuzzy and ill-formed and it still working!

Just:

“Show me invoices”

“Send this to the customer”

And it happens. No searching through menus or remembering how and where to click.

2. The cost of software development is dropping

If ERPs are not what you are interested in, Nikhil showed how to build any software application on the OpenClaw stack. I asked him to build a CRM for the oil industry. He showed me how he:

• Researches the domain

• Drafts a product plan

• Creates the database structures

• Builds functionality

• Deploys it

• Creates a working system

What previously might have taken weeks or months, he showed how to build within an hour.

3. Software is becoming a service (again)

Through our discussion, we landed on a reality - though this all looks easy, organizations will still want to hire people who are proficient in building applications in this way.

This means, instead of buying package software, companies will likely now shift to hiring people (for firms) who understand the domain + the AI + how to build the system.

Building applications in this way is more or less what we - at PreScouter - are shifting towards ourselves.

—

Watch the full demo and my conversation with Nikhil Jathar here:

If you're trying to figure out what “AI-native” means for your company, we’re happy to share what’s working (and what’s breaking).

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

Remember the BlackBerry? Executives couldn't put it down. It was revolutionary: you could, for the first time, email from anywhere. 

And it was so addictive, people called it the “Crackberry”.

Claude Cowork is starting to show similarities to the “Crackberry” - but for all work.

Last week, Anthropic - the maker of Claude Cowork - added a "walkie-talkie" feature. You can be on your commute, at the gym, or at your kid's practice and either text or talk to Claude AI on your phone. Paired with Claude Cowork - running on your computer - it will then follow your instructions and operate your computer.

And, because your computer has access to your apps, your browser and your files, I’ve been able to ask it do do things such as:

• Open the Excel I was reviewing Friday and run that analysis

• Find that market report and answer three questions about it. 

• Start building a slide deck for me using this email

Claude - on your phone - will then share the results back to you.

_{A text message from your phone controlling your computer, as shown in Anthropic’s promotional video.}

This means we no longer need to be tethered to our desks to get work done.

Claude Cowork is doing for work what the BlackBerry did for email.

Microsoft has licensed this technology, so we can expect it to be available in Copilot before the end of the year.

(We’ll have to see how it’s priced and whether it will - like Copilot Chat is to ChatGPT - be a “mixed bag” equivalent using older models). 

But Cowork has a ceiling.

We've been building on OpenClaw - the open-source agent system that NVidia CEO Jensen Huang recently compared to Linux, the foundational operating system running 90%+ of internet servers.

Compared to OpenClaw, Claude CoWork has noticeable gaps:

(1) OpenClaw runs continuously. It doesn't wait for your instructions. You can give it a mission, and it can work in continuous bursts until it gets there. In contrast, Cowork follows your orders.

(2) OpenClaw includes a web server. It can build applications it thinks are necessary, and provide a web interface for you to use them. Cowork can't do that yet.

(3) And OpenClaw is open source. When we’ve hit limitations, we’ve been able to patch the code ourselves.

So while Cowork feels like a BlackBerry, OpenClaw feels like a colleague. 

But OpenClaw’s power comes at a cost.

(a) OpenClaw’s “self initiative” means most people set up a separate machine (or buy a Mac Mini) to use it, so it doesn’t wreck their computer. In contrast, Cowork is safer to use on your everyday computer.

(b) For the best results, OpenClaw needs to use Anthropic’s Opus 4.6 models. API fees for this can run $100+ per day. Cowork is bundled with Opus 4.6 access, maxing out at $200/month. (Our work-around is a complex mix-of-models setup).

(c) Cowork is easy to set up: you simply install an app on your phone and on your computer. Setting up OpenClaw is not as turnkey. 

This makes Claude Cowork the closest thing to a plug-and-play AI agent that exists today.

So the real question isn't which is better. It's which gap closes first.

Does Cowork grow up from a BlackBerry into a platform? 

Or does OpenClaw become more accessible?

We're betting it's both. We're building our “application layer” above OpenClaw to be portable. When Cowork, Microsoft Copilot, or whatever comes next reaches parity, we’ll be able to transfer what we build between them.

The first application we’ve built on OpenClaw is a monitoring agent that tracks topics tailored to highly specific interests - pulling from patents, industry news, journal articles, and more in real time. If you missed our recent demo, you can watch it here:

 AI as a Team Member - Demo

If you’re feeling adventurous, you can try Cowork this weekend. Install the mobile app. Give it a task from your couch. And let me know if you feel that Crackberry feeling… 

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

Just a year after coining the term “vibe coding”, AI leader Andrej Karpathy has again hit the zeitgeist by popularizing another familiar concept with a new frame - coining it “auto-research”. 

If you've felt anxious that you or your organization is falling behind on AI - but couldn't quite articulate what "behind" even means when you're in an industry with real protections - such as physical assets - this post is for you.

🔔 We've built two interactive applications to simulate these ideas. 

See these ideas in action here: AI Optimization Loop Simulations

Before we get to “auto-research”, we need to talk about a bird.

In the 1950s, a biologist named Niko Tinbergen built a fake egg.

It was bigger than a real egg. More colorful. Exaggerated in every way. He placed it next to a mother bird's actual eggs to see what would happen.

She abandoned her real eggs and sat on the fake one.

The bird's instinct was simple: sit on the biggest, most colorful egg. That rule had worked for millions of years as a way of ensuring the survival of her species.

Bigger and brighter really did mean healthier offspring - in nature.

But Tinbergen's plaster egg broke the rule. It was bigger and brighter than anything nature could produce. And the bird's instinct didn't say "wait, that's suspicious." It said "that one. Definitely that one."

Biologists called this a supernormal stimulus - an artificial thing that triggers a stronger response than the real thing it copies.

Every optimization loop at scale has already produced its own fake egg.

You already know social media companies measure engagement: dwell time on each post, clicks, shares. Their algorithms test what keeps you scrolling, keep what works, toss what doesn't. The result? They often push hostile, emotionally charged content. Users don't even prefer the posts shown to them. The platforms optimize for addiction over preference.

But this optimization incentive exists everywhere. 

For example, every company is intrinsically optimizing for repeat purchase - how often you buy again - against what the company sells. Even these loops can have unintended consequences. 

• Ultra-processed foods engineered for addictive repeat consumption. UK childhood obesity has risen 700% in 30 years. 

• Pharmaceutical companies optimizing for prescription volume led to the opioid crisis - drugs were marketed against physician adoption rates, not patient outcomes

• Plastics producers optimized for unit cost and versatility - microplastics are now found in human blood

These unintended consequences often leave companies scrambling for solutions. In fact - some of the projects we undertake at PreScouter are to help clients find new solutions that don’t have these negative side-effects - whether that is carbon emission reduction, sustainable packaging that preserves shelf life or even lower calories.

But, until now, these loops were bottlenecked by supply. 

Social media could only optimize over content that humans uploaded. The food industry could only test recipes that food scientists had time to formulate. Pharma could only test as many compounds as chemists could synthesize. Mining could only evaluate as many site configurations as engineers could model.

That bottleneck is now gone.

An AI agent can generate the supply - the content, the copy, the product variation, the ad creative, the pricing structure, the onboarding flow - and test it. The optimization loop no longer waits for humans to produce the raw material it runs on.

Think about what that means for your industry. 

• A CPG company generating and testing thousands of packaging designs against purchase intent - finding the combination of color, shape, and label copy that maximizes impulse buys regardless of nutritional value.

• A food manufacturer cycling through flavor formulations optimized against craveability scores. 

• An energy company testing rate structures against customer retention while obscuring total cost.

• A chemicals company optimizing safety data sheet language against regulatory pass rates. 

• A life sciences firm testing physician outreach messages against prescription volume.

All these loops just require a goal to optimize against and a machine that can run the loop. This machine - the AI - generates new variations, tests them against the goal, learns from the results and produces the next set of variations in the loop.

Andrej Karpathy’s auto-research is this loop pattern applied to optimizing an AI algorithm: 

• An AI agent creates variations of the algorithm it is improving

• It tests the algorithm’s output against a goal outcome

• Where a variation improves the goal outcome, it uses that as the base for the next set of variations it creates

It’s a continuous optimization loop.

Andrej Karpathy auto-research has the AI optimizing an algorithm towards lower values of the validation metric on the y-axis

For entirely digital loops - such as algorithm optimization, social media feeds, e-commerce ads, email sequences, dynamic pricing, chatbot scripts - the cycle has already been running fast. Our inboxes are full of spam, social media feeds are full of “slop” and even shopping on Amazon can be addictive.

Now, with AI driving supply, the loop accelerates. 

Decades ago, a well-crafted marketing email was a competitive advantage. Now AI generates thousands of variants overnight, and open rates are collapsing industry-wide because every inbox is flooded with competent-but-generic copy..

Value moves from running the loop to designing the loop.

But the pattern doesn't only apply to digital. Evolution itself is an optimization loop that ran for billions of years. Resistance to antibiotics is an optimization loop. Corporate strategy is an optimization loop - just one that runs quarterly instead of hourly.

The difference now is that AI compresses any loop where you can define a goal outcome even when the loop itself is slow. A clinical trial still takes years. A factory run still takes days. A crop cycle still takes a season. AI can't speed up the physics. But it can generate hundreds of variations to test per cycle instead of the three or four a human team would design. 

A manufacturer that used to test five production line configurations per quarter can now test fifty. A life sciences company that used to enter a trial with two formulation candidates can enter with twenty. The clock doesn't change. The number of bets per tick does.

Why “falling behind” is a risk.

Optimization loops coumpound.

The company that figures out how to apply this pattern to the loops that drive their industry doesn't just get a one-time efficiency gain - they get faster at getting faster. Every cycle produces learning that improves the next cycle. A competitor who starts six months later isn't six months behind. They're hundreds of cycles behind. 

And here's what makes this worth getting right rather than getting scared of: the same pattern that produces fake eggs can produce real ones. A drug that works better with fewer side effects is a real egg. A manufacturing process that cuts waste and defects simultaneously is a real egg. A marketing campaign that actually gets people to the value faster is a real egg. The difference is never the loop. It's always the goal metric you measure against.

The question to discuss with your colleagues:

Where could you use this pattern today? Look for processes with clear inputs and measurable outputs. Prioritize those with faster feedback loops or where the cost of creating and testing each variation is coming down. Idea generation. Formulation testing, packaging design, route optimization, supplier qualification, clinical trial design, maintenance scheduling, claims processing, demand forecasting, safety protocol iteration, customer complaint resolution workflows. These are loops waiting to be designed.

Tinbergen's bird didn't make a mistake. Her instincts were perfectly tuned - for a world where nothing could out-optimize nature.

That world is over.

The companies that win the next decade will be those that can create loops that optimize for real eggs.

🔔 We've built two interactive applications to simulate these ideas. 

See these ideas in action here: AI Optimization Loop Simulations

Let us know if we should do more of these types of simulations.

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

We've been building an AI agent powered by OpenClaw at PreScouter over the last few months.

The idea is simple: instead of AI that sits in a browser tab waiting for you to ask it something, what if we had AI that just... worked? On its own. Like a team member would.

It’s not yet a finished product - but I demoed it live last Friday to a few of you in this newsletter community. It was an informal MS Teams call to share where we think things are headed, to get initial impressions.

You can find the recording here:

→ AI as a Team Member (27 min video)

Here are 5 things that seemed to resonate with the room:

1. It lives where you already work.

Most AI tools ask you to go somewhere: Open a new tab. Find the login. Switch contexts. Our AI shows up in your existing communication layer - Google Chat for us, MS Teams for most of our clients - as just another contact. You message it like a colleague. It messages you back.

While this sounds small, it’s a huge psychological shift from an AI tool to an AI as a colleague.

2. It was already working while we slept.

Before the demo even started, the AI had been monitoring packaging industry news overnight. It had categorised findings, flagged the most relevant developments, and pinged me that morning with an update: unprompted.

Even through the demo, it continued to update us on its ongoing work.

We've all used AI reactively - where you ask questions and it answers. This is AI acting on your behalf, with your interests in mind, before you know you needed it.

3. It collapses the tool stack.

During the demo, I asked the AI to find the right contact at a packaging company, search LinkedIn, draft an outreach email, and update the relevant CRM record. 

In our current world, that's four different tools, four different logins, and probably 20 minutes. Our AI rendered a single screen showing all of it - drafts of the work it would do - so I can review and make adjustments, before it updates the relevant tools.

You may have heard of the “SaaS apocalypse” - blunting the value of companies such as Salesforce and Monday.com. This was that in action: One agent can orchestrate across your existing tool stack, collapsing the value of individual SaaS products.

4. It's building its own knowledge - not just searching.

This one is subtle but important: When you ask it something, it's not doing a Google search. It's drawing on a growing knowledge base it's been building since it was first activated - entity relationships, context, connections between things it's observed over time.

We asked it to generate a competitive positioning map. It used everything it had learned since it started running 1.5 weeks earlier - using its accumulated understanding.

5. OpenClaw (and soon Copilot) only provides the blank canvas

Microsoft will soon give you an agentic Copilot. But an agent platform without skills is like hiring someone with a great résumé and no training - they can talk, they can search, they can summarize, but they don't know how your business works.

The value isn't the platform. It's what's loaded onto it: the extraction pipelines that know which industry developments matter to you, the monitoring skills that track your competitive landscape while you sleep, the operational knowledge that turns "we should follow up on that" into an actual follow-up. Platforms are the blank canvas. Skills are the painting. That’s where we are focusing.

The brain in a jar - the point I forgot to make

During the demo, I flashed the picture above - of a person at a desk, next to a brain in a jar (labelled AI) that is operating a computer. For me, this captures the essence of this new generation of AI: a technology that can do anything your or I can do, with access to various data and tools provisioned in the same way we provide those to staff.

But I forgot to drive home the key point:

• Humans live in the real world. We have eyes, experiences, instincts, relationships. 

• The AI doesn't. It’s like an isolated “intelligence living inside a jar,” operating “blindly”.

The implication? This “brain in a jar” scales the thinking of the human sitting next to it.

I don't think this means the end of jobs. I think it means the beginning of something different: the one-person “team”.

An expert in a function - competitive intelligence, sales outreach, market research - can now scale their work, with an AI team member handling the volume.

The human still drives the objectives. This is not a replacement -  it’s leverage.

The practical takeaway

If you're a leader at a large in any organization, after watching this demo, here's what you’ll likely takeaway from the session:

1. Agentic AI is real and it works today - perhaps not perfectly - but well enough to matter.

2. The organisations building the domain-specific layer now will have a head start when the enterprise-grade version arrives.

3. The one-person team isn't a threat. It's an opportunity.

Watch the full demo:

→ AI as a Team Member (27 min video)

Right now, we’re working on implementing this in every part of our organization - with a view to providing this as a service as demand emerges. 

I’m planning to shift this newsletter and related content to document how we’re doing it - the wins, the mistakes and the unexpected moments. I’m also planning to highlight similar work from others, so we can all learn together.

If you have questions or thoughts, let me know, so we can factor those in.

Best,

Dino

Hi, {{FirstName|and happy Tuesday}}.

🔔 We’re demo’ing this on Friday - 6th March, 10am PT / 12 CT / 1pm ET.Reply “INVITE" for the invite / recording.

Last week I got an impromptu chat message from our AI agent - without anyone asking:

"I saw you're working on EU PPWR compliance. 4 companies you're tracking have published compliance statements. Want me to pull up the details?"

Nobody prompted it. Nobody scheduled that message. It just… knew it mattered.

It had been building a knowledge base - automatically - for weeks. It named specific companies, connected them to regulations they'd need to comply with, and flagged a patent filing from 11 days ago that nobody on our team had noticed yet.

Think about how most teams use AI right now.

Someone opens Copilot, asks a question, gets an answer, closes the tab. Tomorrow, the conversation starts again from zero.

It doesn't know what project you're on. It doesn't know that the patent it helped you analyze is connected to the supplier your procurement team is evaluating. 

That's not a team member. That's a search engine with better grammar.

Let me make this concrete.

Imagine you lead an R&D team focused on sustainable materials. Today, staying current means your team manually scanning industry news, tracking regulatory changes across multiple jurisdictions, monitoring competitor patent filings, and trying to connect all of it to your active projects.

That work is important but tedious, and it never feels done.

Now imagine an AI team member configured for your domain. Every morning, it's already processed overnight developments. It messages your team channel:

"A research group at ETH Zurich just published a paper that's directly relevant to the membrane project. And two new patent filings related to mono-material flexible packaging from Amcor and Mondi."

You didn't assign that work. You didn't even know those patents existed yet. But now you do, and you can act on them.

When you ask a follow-up question - "which companies are most active in chemical recycling?" - it doesn't search Google. It answers from structured knowledge it's been building for weeks, with entity relationships, timeline context, and connections to your existing projects.

That's not a chatbot. That's a team member who never sleeps, never forgets, and never loses context.

The 4 things that make AI a team member instead of a tool

We've been building this internally at PreScouter, and a pattern is emerging that is surprisingly simple. There are four capabilities that separate "AI tool" from "AI team member":

1. It knows your world.

Not the whole internet. YOUR world. Your clients, your projects, your team, your competitors, the regulations that matter to you. It builds a living knowledge base from every meeting, every email, every report — and it connects them. When you mention "Acme," it doesn't ask "which Acme?" It knows.

2. It reaches out first.

This is the one that changes everything. Most AI is reactive - you ask, it answers. An AI team member has a heartbeat. It's scanning your domain every day. When something changes - a new regulation, a competitor move, a stalled project - it tells you. Before you knew to ask.

Think about the best analyst you've ever worked with. They didn't wait for you to assign them work. They came to your desk and said "hey, I noticed something you should see." That's what this is, every single day.

3. It does the work, then asks permission.

There's a massive difference between "here's information" and "here's what I think we should do - approve it and I'll execute." An AI team member prepares actions - draft emails, CRM updates, team assignments, research scopes - and presents them for your review. You edit what needs editing, confirm, and it executes.

Need to notify three team members about a project change? It drafts the messages, shows you what it'll send, and waits for your go-ahead. You stay in control. But you're reviewing work, not doing it from scratch. That's the difference between managing and doing.

4. It gets smarter about YOUR organization over time.

Every interaction, every decision, every piece of data makes the system more useful. After a month, it knows your patterns. After three months, it's connecting dots across projects that no individual could track. After six months, it has institutional memory that doesn't walk out the door when someone changes roles.

This is the part that compounds. And it's the part no one-off AI tool can replicate.

Skills are the new apps.

Here's the thing I want to be honest about: Microsoft, Google, and OpenAI are all building AI agent platforms. Within a year or two, the infrastructure layer - the plumbing that connects an AI to your chat tool, gives it memory, and has it be proactive - will be commoditized. 

Within a year or two, what I'm describing will be normalized.

But what those platforms won't build is the skill layer that sits on top:

• What YOU look for in patents.

• YOUR domain configured as a living knowledge base.

• YOUR team's patterns of work learned and refined over decades.

They also won't build the skills you want, but could never easily hire for:

• The mental models regulatory bodies use to evaluate your product.

• The due diligence framework a top materials scientist uses to separate genuine polymer breakthroughs from marketing claims.

• How to audit supplier sustainability claims against actual certifications.

• How a top patent attorney evaluates freedom-to-operate risk in a crowded IP landscape.

• How a world-class venture capitalist evaluates new packaging companies.

That's the layer we're building: the operational knowledge that turns a general-purpose agent platform into a useful team member.

The platforms will give you the engine. We're building the driver who knows where you need to go.

See it live this Friday

We're hosting a 30-minute live demo this Friday: AI as a Team Member (10am PT / 12pm CT / 1pm ET)

I'll be demoing a live instance configured for packaging and materials intelligence. You'll see:

• The agent living inside a chat channel (not a separate app)

• Proactive morning briefings generated from overnight intelligence

• Real-time Q&A where the agent answers from its knowledge base, not the internet

• How we're layering capabilities custom to specific domains

Whether you join live or watch the recording, you'll walk away with a concrete picture of what "AI as a team member" actually means in practice - and a framework for thinking about where this fits in your own organization.

If you'd like to join us, just reply “INVITE” to this email and I'll send over the invite - and the recording for those who can't make it.

Talk soon,

Dino

Hi, {{FirstName|and happy Tuesday}}.

🔔 I'm meeting with some of you on Friday 6th March, 10am PT / 12 CT / 1pm ET, to demo what we’re building with OpenClaw. Reply “INVITE" if you want to be added to the invite or to get the recording.

On a weekend this past November, Peter Steinberger opened his laptop and started tinkering.

He had a simple thought: what if AI could actually do things? Not just answer questions, but take action?

Book the flight. Fill out the form. Update the CRM. Join the meeting. Handle the thing.

Though the name required a few iterations, OpenClaw was born.

Then the internet found it. And everything changed.

Within weeks, it became the fastest-growing GitHub project ever.

(Github is a platform that hosts software projects, allowing anyone to contribute to them).

More tech-inclined business operators immediately recognized what they were looking at: A new kind of employee.

One that doesn't need dopamine hits to stay motivated, a ping-pong table, a quarterly review, or a 1:1 to feel heard. One that just... works: Clicking buttons, filling forms, navigating software - and doing it all without burning out, rest room breaks, or morale boosts.

Peter Steinberger recently spent a week in San Francisco meeting with every major AI company.

On February 15th, Sam Altman, CEO of OpenAI, posted:

No compensation terms were disclosed - but OpenAI recently paid over $6 billion for Jony Ive's startup. They are not writing modest checks.

One person. One weekend project. Three months. A life-changing outcome.

This is not a story about AI replacing humans.

It’s a story about leverage. It shows that one person with good judgment plus AI tools can create large amounts of value fast.

The value that OpenClaw created - and why it has been a sensation - is because OpenClaw itself creates leverage for the rest of us.

We can apply the lessons from this story to how we use AI, through tools such as OpenClaw, to our own work.

1. Pick a leverage point, not a pain-point. Peter didn’t build to solve a singular pain point. He went after a leverage point: “What if one person could coordinate a whole swarm of digital workers that click, type, and handle your computer for you?”

We can apply the same type of thinking to our own work. Where in your world does a single improvement unlock 10x leverage across everything else? Is it:

• Agents that watch your inbox and systems.

• Tools that cut away entire layers of coordination and data entry.

• Systems that turn institutional knowledge into something searchable and reusable.

2. Use AI to replace tasks, not expertise. The magic wasn’t “AI did everything.” AI handled the repetitive parts; Peter supplied the expert judgment - what the agent should do, what good UX feels like, and what constraints matter in the real world.

Two questions for us to keep handy:

• What am I still doing that a smart agent could do 90% as well?

• Where does my judgment, domain knowledge, or expertise actually matter?

AI can tackle the first list, but we need to staying glued to the second.

3. Ship to the network, not the org chart.  OpenClaw didn’t roll out through persuading stakeholders; it spread through GitHub, dev chats, and social feeds.instagram.

We can ask:

• Can a single user get value in 10 minutes?

• Can they share it with someone else in one click?

• Does using it make them look smart or early to their peers?

4. Think “AI agents & humans,” not “humans vs AI agents”. The endgame here isn’t replacing everyone. It’s about how to redeploy them - moving humans up the value chain and letting AI handle the work that was never a great use of human potential to begin with – the work that requires judgment, expertise, relationships, and genuine creativity.

Which brings us to what we're building.

What we're doing with all of this at PreScouter.

A few weeks ago, we set up OpenClaw and quickly realized that the “out of the box” configuration wouldn’t work for us, because it’s designed as a single user experience.

As such, most people setting up OpenClaw are setting it up as an assistant - an AI that helps you with your personal tasks, such as writing emails or making calendar bookings.

For us, this has some value, but not enough value.

We’ve instead been working on setting it up as a caretaker - an AI employee whose job is to tend to the operational infrastructure of our business.

Our team manages a huge amount of “metadata” - client names, deadlines, assignments, Salesforce updates. These are all necessary, tedious and a quiet tax on hours that could go toward client work. We're building OpenClaw to handle this, so our people handle what they're actually at PreScouter for.

But that's just the beginning. The longer-term vision is an always-on AI employee that soaks up documents, transcripts and other knowledge to function as institutional memory. When someone leaves, their knowledge doesn't leave with them. 

We could ask "Why did we recommend that technology to this client?" and actually get an answer.

We're building something that will change how our company operates and remembers.

Want to see it in action?

🔔 I'm meeting with some of you on Friday 6th March, 10am PT / 12 CT / 1pm ET, to demo what we’re up to. Reply “INVITE" if you want to be added to the invite or to get the recording.

Dino