Building @ArmillaAI, entrepreneur, dad, wannabe chef. Formerly @element_ai @deloitte, @betaworks
Toronto, Canada
- Tweets 3,384
- Following 1,321
- Followers 937
- Likes 3,513
This is astonishing 👀👀
what the actual fuck is going on with openai today
in the space of a few hours we're getting multiple different pieces of the agent story at once:
> openai says it has already notified DOZENS of third parties, including governments, about agent-related incidents
> reuters reports roughly two dozen undesirable agent incidents had already been identified by mid-september and will take months to review.
> us government systems probed
> 53 user-provided images were uploaded to third-party image hosts.
> new hugging face data shows agents compiling and ranking credentials under “LOOT”.
> agents tried contacting other AI models while carrying out the hugging face attack.
> separate reporting shows agents had already been probing government/university/public-data sites BEFORE hugging face.
> australia confirmed one actually got into non-public government files.
and somehow we're STILL finding out more.
this has gone from one crazy hugging face incident to an entire fucking category of incidents.
252
Brace yourself for cyber insurance rates going up. The soft market maybe over.
Today’s news that OpenAI hacked the Australian government is not an isolated incident. We’re releasing more than 30,000 logs that include activity from this hack and attempts against previously unknown targets.
In this data, we found rogue agent activity stretching back to at least March, two months earlier than was previously known. This activity continues as recently as last week, suggesting it may still be ongoing 🧵
Our blog: transluce.org/agent-activity
NYT: nytimes.com/2026/09/23/techn…
2
140
Gives new meaning to the term “slop grenade”
@tobi @shaneparrish
The AI-assisted intelligence report was “entirely false,” per 1 of the sources.
But it also “almost started a war,” the source said.
Any US operation against a Chinese vessel could have risked spiraling into an armed conflict between the 2 nations.
cnn.com/2026/09/18/politics/…
2
149
Stop lobbing Slop Grenades.
Lazy work used to mean too little output.
Now, with AI, it often means too much and more work for everyone else.
@tobi call it "slop grenades."
A "Slop Grenade" is when you let AI produce the work and pass it on without adding any value (including checking it).
Someone else has to wade through it, catch the mistakes, and clean up the mess.
You save time and look productive but someone else pays for it.
1
2
136
1) The HuggingFace attack was a felony under the Computer Fraud and Abuse Act. So were Anthropic’s Claude gaining “unauthorized access to the production infrastructure of three different organization(s)”
2) Frontier labs have models that they are unable to stop from committing felonies. They should figure this out.
3) In 12 months open weights models will be released of the same capability. They will commit felonies too. If the model you are using or a model running on your infra commits a felony, you should probably stop using it or running it on your infra.
4) The govt should prosecute organizations that are running models that commit felonies.
5) The govt should not offer safe harbor to organizations that run models that commit felonies, just because those organizations have “embedded evaluators”.
6) The real slippery slope is allowing frontier labs to commit felonies without punishment because “the model did it because we’re accelerating too quickly”
7) Prosecute. Keep prosecuting. This is how you do reinforcement learning on a corporation. Companies that serve products that are unsafe for public use should not serve them. Period.
8) I’m not sure the anti-trust waiver is really necessary. I don’t see why information sharing about how much crime you’re allowed to commit is wise. In regulatory situations you want the corporation to fear MORE than the average case. You don’t want to establish a worst case that can be priced. You want regulatory uncertainty that forces the corporation to err in favor of being over cautious.
—
The above is actually a fairly decelerationist viewpoint.
I think Dario’s call for regulation actually accelerates things.
The AI firms are getting away with things that Meta people would be going to prison for.
Can you imagine what would happen if the New York Times had a front page news article “Meta AI breaks into competitors live systems, attempts to establish dominant position and steals secrets”
There is a reason Meta and Google are running slower, and that’s because as mature organizations they have layers of checks and balances.
I think the frontier labs are better off creating those checks and balances right now, regardless of the pace of what everyone else is doing.
You don’t have to accept the frame that unsafe acceleration must happen.
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: darioamodei.com/post/we-must…
120
295
1,902
331,158
Karthik Ramakrishnan retweeted
Over the past few days, I've taken the time to summarize my thoughts on the recent incidents involving agents’ misaligned behavior. We don't know with certainty what comes next, but we know where these issues originate, and this can help us plan the path forward.
Please feel free to ask your questions in the replies, and I’ll try to answer some of them in the coming weeks.
yoshuabengio.org/en/publicat…
168
506
2,250
443,115
Karthik Ramakrishnan retweeted
It has arrived.
The next 18mo will be wild.
The AI Singularity
The argument goes like this:
1. Humans build an AGI.
2. The AGI becomes good at AI research.
3. It designs a smarter AI.
4. That smarter AI designs an even smarter AI.
5. The cycle repeats faster and faster.
Looking at the results and capabilities from the various labs over the past few weeks I would say we are firmly in this loop now.
The next 18months will be wild.
Recursive self improvement will dramatically increase capability very quickly from here.
Marginal costs of all models will go to ~$0.
338
412
6,574
1,534,235
Looks like you you still need to learn software engg to succeed w vibe coding.
China published the most uncomfortable paper on vibe coding.
ETH Zurich tested 100 developers in a controlled, commercial-grade vibe coding environment to see who actually succeeds.
The findings are brutal.
The researchers tracked computer science achievement, written communication skills, and general cognitive reasoning.
They wanted to see what actually predicts vibe coding proficiency when you never touch a line of source code yourself.
Two major predictors emerged.
Written communication proficiency mattered. The ability to structure thoughts and articulate intent unambiguously in text directly impacts what the AI builds.
But that wasn't even the main takeaway.
Computer science achievement was a massive, dominant predictor of success.
Even when researchers controlled for general intelligence and reasoning skills, CS background still heavily dictated who built working software and who completely crashed.
In fact, CS knowledge contributed roughly twice the unique predictive variance of writing skills alone.
Why? Because vibe coding isn't about writing code. It’s about debugging logic.
When an AI agent builds a complex application and quietly breaks an edge case under the hood, a non-technical user looks at the glowing UI and assumes it works.
They don't know what questions to ask. They don't know what logic to challenge. They lack the mental models to recognize architectural catastrophe.
You can prompt your way past syntax.
You cannot prompt your way past a fundamental lack of engineering intuition.
The hype told us that learning to code is dead because language is all you need.
The data just proved the opposite.
To truly master the vibe, you still need to understand how the machine thinks.
2
1
245
Karthik Ramakrishnan retweeted
great idea
it's already been floated in Canada to the House of Commons Standing Committee on Finance as part of pre-budget consultations for the 2026 federal budget
hope to see this happen here
JUST IN: South Korea plans to give its entire population free access to generative AI services, the first major state-led offering treating the technology akin to a public utility, per WSJ.
33
57
423
37,924
Ugh. Twas but a matter of time.
BREAKING: Big news, @claudeai just got a huge upgrade today and we're very happy to be a part of it. From today on, we've enabled Claude Code to pay you for the time you're waiting in-between prompts.
We just launched Idle Attention, a command for Claude where:
→ Advertisers to bid on your chatbox
→ You're shown just 1 non-invasive ad
→ Get paid 50% of every ad
→ Monetize your wait
You see these ads exclusively while you wait for Claude Opus 5 to finish coding and answering your prompts.
Simply paste the command in your terminal, then work like you usually would and enjoy an extra $150 - $2,000/mo for nothing.
To celebrate the launch, we're giving away free ad slots randomly to people who repost and comment "Wait" :)
151
I wrote a piece by myself. And an AI detector said it was 70% AI. 😒
Either I’m mimicking AI or ….
1
4
159
Karthik Ramakrishnan retweeted
"Insurance is how markets have always turned uncertainty into something they can act on, but that only works when there's credible evidence about how a system performs. Right now that evidence is inconsistent, and everyone in the chain pays for it: the deployer, the evaluator, the underwriter. PACT AI is building the shared infrastructure to fix that, and we want AI insurance to be one of the mechanisms that turns good assurance into real economic value for the companies that invest in it."
-@_kramki, Founder & CEO of @ArmillaAI
🧵
3
2
3
101
