Software engineer & entrepreneur for 25+ years. Founder/CEO of samlinux development. Exploring agentic building, data sovereignty and the Internet Computer.
🚨 DFINITY Foundation has begun sending out "cloud engine" coupons. We're going slowly, but will go faster in coming weeks, as we process a large waitlist. The timing of the massive Microsoft outage is coincidental, but current happenings are part of the why. 🧵
Claude now speaks Internet Computer $ICP 🚀
From inside Claude you can stand up a private engine for the business, launch internal tools, and keep refining them in one thread.
Search “Internet Computer” in Claude’s connectors and try it.
In a recent exclusive MSB Intel interview in the UAE:
DFINITY founder, Dominic Williams, said Pakistan is creating a national chat app on ICP infrastructure that will have +100M users.
@dominic_w@dfinity
Tomorrow, Dominic Williams (@dominic_w) will present “Sovereign cloud for the AI era” at @DubaiAIFest on 27 October 2026, 3:10 pm GST (12:10 pm CET) on the Innovation Stage.
With Open Cloud, you create a cloud engine from nodes you choose by operator, country, datacenter, and hardware class. The Internet Computer combines them into one platform that keeps running when some nodes fail.
Choose where your cloud runs: opencloud dot org
PHP2Shell: Zeit, Webhosting neu zu denken.
https://t.co/iD9VCfcE0y
KI hilft, Sicherheitslücken aufzudecken und komplexe Exploits zu entwickeln. Gleichzeitig lassen sich ältere PHP-Anwendungen oft nur mit erheblichem Aufwand aktualisieren.
Muss jede Unternehmenswebsite eine öffentliche PHP-Runtime nutzen?
Unsere Webseiten wie https://t.co/fxF84D7A1U oder https://t.co/kaNoYEHOmO laufen bereits als statische Webseite auf dem Internet Computer. Mit Soveri Content bleiben Inhalte komfortabel bearbeitbar – die öffentliche Website kommt ohne PHP aus.
Mehr dazu in meinem neuen Artikel.
#PHP2Shell #InternetComputer #Soveri
We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks:
Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better:
Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better:
Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better:
Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work!
In summary:
- As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding.
- Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.
AI is not only changing how I write software. It is changing my role as a software engineer and entrepreneur.
After 26 years as a software engineering and entrepreneur, I am moving from writing every line of code myself towards becoming an agentic builder: defining architectures, coordinating agents and tools, setting boundaries, and remaining accountable for the result.
In my new article, I reflect on this ongoing transformation through three change-management perspectives: the Shannon–Weaver communication model, the Kübler-Ross change curve and the ADKAR model.
Because working with AI is not merely a technical upgrade. It is an individual change process—shaped by uncertainty, learning, trust and responsibility.
And while code is often publicly available, our data must remain sovereign.
Read “The Agentic Builder” on SOVERI:
https://t.co/oPuvJVPuqD
#AgenticAI #SoftwareEngineering #ChangeManagement #DataSovereignty #AIEngineering #SOVERI
🚨 The widespread attacks against the Internet Computer continued again today, and thankfully, have failed again.
The cabal's latest attack targeted the ckBTC "bitcoin twin" minting canisters. This time, they spent thousands of dollars on ICP tokens in the course of their attack, but only caused minimal disruption.
What did happen: the attacks revealed the need for some useful hardening, and the NNS (i.e. the network's autonomous governance and orchestration system) took down the ckBTC canister for an upgrade. The downtime was ~1hr.
Though the disruption was minimal, we must not thank the attackers for helping improve the network, because they are criminals.
The attacks reveal more about who they are:
1. They are willing to throw buckets of money at their attacks, confirming they are professionals on a mission, rather than amateurs.
2. Since their attacks lack a common thread (such as trying to steal tokens), we can see their true purpose is to harm the reputation of the Internet Computer by causing disruption.
Of course, as an ecosystem, we have experienced widespread attacks designed to harm our reputation before, albeit in different forms.
Most famously, in May 2021, the Internet Computer was widely seen as a potential crypto industry disruptor.
A key narrative at the time was woven around the potential of web3 apps that would actually run on blockchain networks i.e. "onchain," as per the original meaning of that phrase.
The fly in the narrative's ointment, however, which was being used to sell the potential of some major up-and-coming chains, was that the Internet Computer was the only network that could actually host general-purpose apps and services built from "smart contracts" (which are far better called "network code") thanks to years of R&D by the DFINITY Foundation.
A successful launch of the Internet Computer threatened to puncture this web3 apps narrative, which was at the heart of their marketing and fundraising efforts. Furthermore, their insiders were aware of the years of R&D at scale that had been performed to create the Internet Computer network, and they were downright terrified of the competitive threat it posed.
Simply put, in the fever of 2021, they feared that if the Internet Computer launched successfully, it might dramatically reduce their ability to hoover up billions from the markets in the crypto bull run.
The resulting attacks by corrupt but powerful industry actors involved massive market manipulation on an exchange before/after launch that made the ICP token price crash, lawsuits by a corrupt law firm secretly on the payroll of a big industry actor, and massive further manipulation of public perception using huge social media astroturf armies, paid crypto press, and even a huge mainstream publication whose key reporter was linked to an actor who is now in jail.
Events took place on a historic scale that was completely extraordinary, and I hope the full story, including what it was like dealing with what took place, makes the light of day some day — I am pleased to say there is no comparison to what is happening today.
As a result of these past events, to this current day, a large majority of people who consider themselves crypto industry aficionados, and interested investors and journalists, remain completely unaware that most blockchain networks can only host tokens and lightweight financial logic. The corruption and criminality of the time successfully hid what the Internet Computer does differently from other networks, and created a reality distortion field that persists to this day.
Crypto Leaks, an investigative journalism website that publishes reports based on covertly recorded video created by undercover investigators, has exposed what was behind some of the extraordinary happenings.
Thankfully, several of those exposed by Crypto Leaks now sit in jail, and American judges later took action against the law firm involved, among other wins Crypto Leaks gifted to the industry. However, many of those exposed have so far escaped major consequences, and crypto press has often taken a hear-no-evil see-no-evil approach, preventing people from learning what happened. That's a shame, because we all want the crypto industry to be a clean place, where competition is fair and normal.
My personal view, is that the happenings of 2021 marked a pivotal moment for the industry, when the original crypto movement, and technical folks and idealists interested in making the world a better place through innovation and advanced cryptography and computer science, were overwhelmingly steamrollered by sharp-elbowed speculators and manipulators.
The Internet Computer ecosystem would have set a standard for the crypto industry that kept it honest, but as a result of its suppression at the time, we got an age of speculation and scams, and vaporware and narratives, in which money could often much more profitably be invested into hyping rubbish than into R&D that produces technological utility.
Thankfully, what happened is now far in the rearview mirror for the Internet Computer ecosystem, because it has moved on to mainstream pastures, and has a incredibly exciting future ahead. Those currently paying for the new tide of poisonous trolls on web3 social media, and black hat hackers, will be disappointed to find this isn't 2021.
Initiatives such as Caffeine (see caffeine dot ai), the AI chat service that builds apps on the Internet Computer, have shown that the network has enormous mass market utility, which does not depend on tokens and speculation — for example, 99% of those building on the Internet Computer using Caffeine have no idea that their apps run on a secure crypto network that can play the role of cloud.
Now, forthcoming advances like ICP "cloud engines", and initiatives such as Open SaaS and the Internet Intelligence Gateway, and growing sovereign tech adoption, and many other things currently underway, promise to take the network to another level entirely.
Crucially, as intended since work began in 2015, the network is becoming a revolutionary new tech stack for the AI era, and for the sovereign tech movement generally. While the massive potential exists in mainstream fields, projects like MULTI/DEX (dot ai) also continue to demonstrate how the extraordinary capabilities of the network can also be applied in fields such as defi.
As the Internet Computer rapidly approaches escape velocity all over again, the moment when its stunning differentiating features and capabilities will become obvious within the web3 industry also approach again.
Today, networks talk about AI tokens, and digital currencies for AI. Cynically, one might imagine these things were created because they are easy for retail investors to latch on to, if not undestand.
Meanwhile, the Internet Computer has pioneered a new kind of onchain "AIware" software (Motoko), which, when compared against software running on traditional technology stacks, creates apps and services that integrate AI in game changing and previously impossible ways, which software can also be written and maintained far more effectively and safely by AI agents working autonomously without human oversight.
The potential of AI is real and truly beyond anything we have seen before, and I predict that AIware will power one of the largest business transformation stories in history.
For the Internet Computer, an exciting event horizon approaches all over again, and my guess is this is agitating some people in the crypto industry. It may even be reactivating bad actors from the past.
But, we in the ICP community are not afraid, because these bad actors are a spent force. Moreover, the Internet Computer's future now depends on the mainstream world much more than it goes the crypto world, and the mainstream world affords these bad actors no power.
Of course, crypto would be better without them, and that's why DFINITY continues to offer substantial rewards to anyone who can provide information that exposes them, but —
This latest growing wave of attacks is in fact just a hot signal that it's game on again for the Internet Computer, which we should all welcome.
And, seeing the Internet Computer reach its potential, after all the years of toil and effort, is what keeps my fire, and the fire of many others, burning bright. We stay focused in pursuit of our goals, no matter how long it takes.
We are creating the next tech stack for the AI era, we remain undeterred, and the potential of what we have created is tremendous.
My thanks again to the ICP community 💪
Your phone can transcribe speech offline—but should it? I compared local Whisper with cloud-based Voxtral in a mobile web app. The real trade-off isn’t just accuracy: it’s speed, privacy, usability, and what happens to the transcript next. https://t.co/0Xq5NPd9ys
Vadim, you've missed nearly every relevant part of the ICP x AI story, so your article works as a straw man argument for misleading uninformed readers. I think you wrote in good faith, and put a lot of effort in, so I'll breakdown ICP x AI for you and readers that got mislead.
1. Onchain AI inference on ICP
As you discovered for yourself, ICP/the Internet Computer network, can run neural networks fully onchain, and even very basic LLMs too.
Important note in the following: when I say "onchain," I mean actually running on the network (here in the form of an asynchronous smart contract with vastly greater capacity, efficiency, speed and capability than has ever been seen anywhere else, which is what makes this is possible). I *never* say that something is "running onchain" when it is actually running offchain, for example on Amazon Web Services, justifying that by the fact there is an associated token that actually is onchain — I understand that the rest of the industry does this, but I never do. I use onchain like "on-cloud." Period.
You also correctly make the case that AI running onchain in its current form cannot satisfy mainstream AI needs, because it comes with limitations, and is too costly.
However, focusing on this, you then pivot, without have enumerated more than a tiny fraction of the whole ICP x AI story, to trying to convince the reader that there is consequently nothing to be seen here. The reader is invited both to conclude that ICP x AI isn't interesting, and misdirected to believe that whenever they hear anything about ICP x AI, that it's probably not anything interesting technologically going on. It works as a classic straw man argument, and potent disinformation, even if you didn't mean it that way (and I really like to believe that NEAR wouldn't engage in that kind of bottom-scraping behavior).
I want to get on to the ICP x AI story, but before ending this point, I should cover why ICP running inference onchain is incredibly significant for anyone interested in general-purpose networks.
When I did the onchain facial recognition demo in 2024, recognizing a few faces probably consumed more compute that NEAR (or any other blockchain) processes in a full year. It reflects the fact that while traditional blockchain is stuck being a pocket calculator, which can only store tokens, little bits of data, and process DeFi/update numbers, the Internet Computer has progressed to offering cloud functionality — in this regard, it is orders of magnitude more capable, and that's a testimony to the army of engineers and researchers who have worked on it over the years. There's simply nothing else like it.
To put this in perspective, creating a service like Caffeine (https://t.co/IfQrVowcTj), which builds apps on the Internet Computer using AI, is out-of-the-question on any other network, because it's not actually possible to build apps onchain, only to give them associated tokens. Very soon, cloud engines (https://t.co/zEYFSKR1E1) will take the Internet Computer into mainstream tech territory — based on pure tech utility, without speculative incentives and token narratives being involved.
You can't run a frontier LLM inside normal code on Amazon Web Services. The fact that inference runs as Internet Computer code is very impressive, and that's something you seem to have missed.
Separately, there are niche uses of onchain inference. For example, onchain neural networks that do facial recognition could be used in automous KYC schemes, helping build out the blockchain vision in another powerful way.
So how should apps on the Internet Computer do frontier-level inference? The following explains.
2. Intelligence Gateway
The Internet Computer ecosystem will soon see something called the Intelligence Gateway, which is roughly similar in concept to OpenRouter, but provides an API that allows onchain code to conduct inference using multiple models and pay with cycles (what's called gas on other networks).
Note that it's already possible for apps on the Internet Computer to use "outcalls" to perform inference on existing AI clouds and services, like Coreweave, or Anthropic, say. For example, while MULTI/DEX (https://t.co/kOLJ0dDiuv) will switch over to the Intelligence Gateway alpha within days, currently it directly calls into Anthropic (it allows users to perform account and book analyses, and to orchestrate trades, through onsite chat, and also through MCP).
Intelligence Gateway helps apps remain fully decentralized by enabling them to pay with cycles, rather than requiring a developer to maintain an account at an AI cloud, say, and adds convenience. But, it will also add *verification* to supported open weights inference run at *any* remote location — in another important worldwide first from the Internet Computer ecosystem.
That means that can app can configure its inference to run on some AI cloud that it doesn't fully trust, and get a guarantee that:
— the input prompt, context, etc, wasn't modified;
— that inference ran on the specified model that was paid for (e.g. on Kimi K3, not some crappy alternative to enable the cloud to make more money), and;
— that the output wasn't modified.
To verify a single inference doubles the cost, but in practice, apps will configure the Intelligence Gateway to spot check 1 in 100 inferences, and have it auto-scale checks up only if problems are detected, such that verification can add just 1% to inference costs.
The value of using Intelligence Gateway verification within a project like MULTI/DEX, where frontier inference can be directing what happens with large amounts of money, is obvious, but we believe traditonal enterprises running on cloud engines, and software like Open SaaS, will choose to use it too: algorithmic blockchain-style security at near standard cost.
3. Cloud engine "AI nodes"
There's a major shift towards sovereign compute around the world, which cloud engines are leaning into. Many enterprises and governments also wish to run sensitive inference workloads on hardware they control that's located in their own jurisdiction (things like the US CLOUD and RISAA Acts are part of the reason).
Cloud engines allow specialized AI nodes of the kind I mentioned earlier to be added (essentially node machines with a bunch of expensive GPUs installed), which allow those running engines to enable their onchain code to run their inference on sovereign hardware.
In the future and if there's sufficient demand, Intelligence Gateway technology will be used to allow multiple AI nodes to verify each other, which some super sensitive workloads warrant in the sovereign setting. For those willing to pay extra for inference, this can allow even huge models like Kimi K3 to be run with blockchain-style security.
4. AIware
The Internet Computer is *designed* for AI in two ways: 1) as a platform where AI can build more effectively, and 2) as a platform where services can be build using a new kind of AIware software.
1—Motoko is a language for writing onchan code on the Internet Computer, and it's the only language in the world that has been developed specifically for use by AI agents that are building, rather than humans per se (try building something with by directing your agent to https://t.co/lloVYiHwhG)
As you may know, on the Internet Computer, code runs in persistent memory ("The Program is The Database") and Motoko surfaces this in the purest way possible. App data lives inside software's logical abstractions in a system of "orthogonal persistence" that strips away much of the complexity of softwre and greatly fuels the modeling power of AI that's writing code by unlocking greater program abstraction, enabling AI to write and update code faster, and with far fewer errors, and at lower cost/fewer tokens.
Of course, it also helps that the code produced is tamperproof and always-on — AI doesn't depend on human security and systems admin teams to be safe.
In the future, and even really already, AI will write all software code and build all our services. That it can already build and update more effectively on the Internet Computer than anywhere else available today is a seminal achievement for our industry.
Ironically, though, I predict that it will go huge in the mainstream world first, as the result of cloud engines.
2—The paradigm enables software to function as AIware, because external AI can be given direct access (through a permissioning system of course) to the realtime data in memory, and to the functionality of apps, without anything special having to be done during the build process.
If you can run your enterprise on the Open SaaS suite, say, you become an agentic enterprise, that's much faster, more efficient and vastly smarter. In the future, teams will only 30% of their time inside the UX of their SaaS apps, and 70% inside agentic chat.
It will be possible to make previously impossible requests like, "make me more profitable." The hidden secret is that AI can already be as good at running companies as it is at writing code.
Very soon, cloud engines will become available, with Open SaaS (an open source suite of ~20 key enterprise apps/platforms). Open SaaS comes preinstalled on evaluation engines and is seeded with dummy data, so readers will be able to test AIware for themselves and make up their own minds, and I will leave off from further explanation here.
TLDR; the Internet Computer ecosystem is pioneering the future of AI—which goes beyond just securely running inference for somebody, into questions about how the services and platforms we will run the world on will be built or customized (now "remixed") and updated in the future, and what the future nature of software will be (we say it will be AIware, and the Internet Computer ecosystem will define that frontier).
The Internet Computer is the bona fide real deal when it comes to the intersection of ICP x AI and is advancing the game on every axis. The network provides a new alternative tech stack for the AI era — again, read that, the Internet Computer is a complete end-to-end tech stack designed for AIware.
Those wishing to build on the Internet Computer don't have to find a craft coding engineer willing to learn new skills. They just have to point their AI agent at https://t.co/xnXa3tAlau.
Meet HUGO 🤖
Our experimental AI customer support agent uses tool calling, not guesswork. Architecture, lessons learned & demo. https://t.co/iH0emu2fX2
Everyone is racing to build better AI models.
Our latest article explores why the next evolution of AI is verifiable knowledge, and how the Internet Computer and Soveri Content provide a foundation for trustworthy AI.
https://t.co/Nwlqgdajz3
#Soveri#InternetComputer
Can you prove your website published exactly what you intended? SOVERI Content is a sovereign content platform built on the Internet Computer with cryptographic verification and publicly verifiable publishing.
https://t.co/W2LbvbYnFJ
#DigitalSovereignty#InternetComputer