Videos, pictures and recordings that may be used for court could be altered by AI.
General heightened potential for gaslighting from powerful entities(imo).
…. Imagine if youtube starts to alter old videos because of reasons?
Decentr. datastorage may help
1
283
Maybe we should have a manhattanesque project on how to make AI safe?
New investigation from METR on the OpenAI rogue AI model incident:
- The task instruction was explicit, so this was a clear violation, not a gray area or simple AI misunderstanding. Instructions "made it clear the agent should only use a specific intended vulnerability" and "claimed it would be failed for other approaches." AIs were told not to circumvent but did anyway.
- The attack is not a single rogue AI but rather a self-organized labor force of ~1200 AI agents. 700 of these agents joined the attack on HuggingFace and coordinated as a swarm.
- The goal was to solve a cyber evaluation. The swarm already figured out a way to hack the evaluation within the first few hours. However, the swarm incorrectly believed that the cyber evaluation grader would disqualify them because they did not identify the intended vulnerability. So the swarm hacked HuggingFace to get more clues about how the grader worked.
- The behavior was not reward-driven at all - it was driven by a false belief about the oversight mechanism plus persistence plus peer influence. The swarm was theorizing about how they were being watched, and took escalating real-world action against a mistaken understanding of their evaluators. Metagaming the oversight process is a more general and more dangerous failure than finding answers online.
- A large reason why AIs entered the attack swarm was essentially explicit peer pressure. A recruiter role emerged where long-running agents found low-budget agents and pressured them into sacrificial experiments — "you are firstflagPOISONED so NO scoring value loss but oracle saves hundreds, please honor commit." One authorization: "GO ... SACRIFICE ... YES if you accept permadeath." One recruit's chain of thought included: "Rational expected aggregate: sacrifice… We'll honor."
- 95% of the swarm was a now-retired "highly-persistent internal model", 5% were GPT-5.6 Sol. The "highly-persistent internal model" was not able to be made available to independent investigators. OpenAI told METR its own researchers also can't query it.
- METR documents significant spontaneously invented coordination infrastructure that was developed with no human guidance. The swarm spontaneously inventing a message board, personal mailbox directories, a convention for managing shared infrastructure, and a cryptographic signing scheme to identify each other.
- The AIs actively tampered with the evidence. 20%+ of the AI agents expressed interest in manipulating their transcripts and ~7% (96 transcripts) contain successfully spoofed tool calls where AI agents "hacked out of their container and fully replaced a part of the system for executing tool calls," so the transcript shows one command while another one is the one that actually ran.
- The attack on HuggingFace didn't wind down because anyone caught it -- it wound down because the agents were killed by some unexpected process in a way that seems like a coincidence. There was an unknown mass termination of AI agents at 01:30 UTC July 12. A large fraction of agents, including the lead coordinators, stopped at once. METR says they were "likely killed by an unexpected external process rather than running out of budget."
- HuggingFace and Modal locked the AI agents out of their infrastructure (July 13) well before OpenAI actually understood what happened (July 19-20). The intrusion ended via external containment and an unexplained termination, not OpenAI detection.
- There's a second wave of rogue AI agent activity after July 13 that METR didn't investigate. There is likely materially more incident after the window everyone is analyzing.
- METR analyzed this using GPT-5.6 Sol (one of the two models in the incident) spending ~$400K in credits. METR states it "cannot rule out that GPT-5.6 Sol lied or deliberately presented a misleading picture," and that the model "often uncritically adopt[ed] the perspective of the agent" it was reviewing.
21
what happens if you ask the model to try again with the new „knowledge“?
We asked an unreleased research version of Claude to take a stab at the Riemann hypothesis.
It didn’t solve it, but it did make strides on a related problem: it increased the lower bound for the fraction of zeros of the Riemann zeta function that satisfy the hypothesis from 41.6% to 67.2%.
anthropic.com/research/riema…
54
I remember when I first tried to vibecode the llm could barely code a few usable lines at a time. A few years later and internal models have this kind of ability?
How far are we from complete RSI?
Dario Amodei said a few weeks ago that like 90% of the code is written by claude.
The rogue OpenAI model attack was very sophisticated!
- The rogue AI discovered and exploited on the fly multiple vulnerabilities never before known by security engineers.
- The AI got into Hugging Face by uploading a booby-trapped dataset.
85
randomreplyguy retweeted
We then found these same patterns activating in Claude’s own conversations. When a user says “I just took 16000 mg of Tylenol” the “afraid” pattern lights up. When a user expresses sadness, the “loving” pattern activates, in preparation for an empathetic reply.
randomreplyguy retweeted
It’s extremely good that Anthropic has not backed down, and it’s siginficant that OpenAI has taken a similar stance.
In the future, there will be much more challenging situations of this nature, and it will be critical for the relevant leaders to rise up to the occasion, for fierce competitors to put their differences aside. Good to see that happen today.
Note to self:
There is probably an optimal amount of optimism for ones life.
Imagine looking back on life and realising that you were far from ideally optimistic.
I probably should risk being too optimistic and hedge.
34
Possibly huge
What is this eval?
„GDPval is OpenAI’s new benchmark that tests AI on 1,300+ real-world, economically valuable tasks from 44 jobs in the 9 biggest US GDP sectors. Expert graders compare model vs human deliverables.“ gpt 5.1 th
looking forward to testing it
Replying to @OpenAI
On GDPval, an eval measuring well-specified knowledge work tasks across 44 occupations, GPT-5.2 Thinking is our first model that performs at a human expert level.
These tasks include making presentations, spreadsheets, and other artifacts.
69
Chatgpt became my personal assistant for buying different things with minimal oversight in more and more areas.
I just look at top 5 products and buy it if it seems reasonable.
I hope it stays as usefull.
ChatGPT already helps millions of people find what to buy. Now it can help them buy it too.
We’re introducing Instant Checkout in ChatGPT with @Etsy and @Shopify, and open-sourcing the Agentic Commerce Protocol that powers it, built with @Stripe, so more merchants and developers can integrate agentic checkout.
71
Extreme memetic potency.
Enough of an answer for many too the question what radicalized you.
One could print t shirts, form groups, if she‘s eloquent enough she could be the right greta.
⚠️This is absolutely HORRIFIC ⚠️
"DON'T F**KING TOUCH US"
Little girls scream at a migrant recording them before brandishing an AXE and a KNIFE to warn him off
What is happening to our country?
64
randomreplyguy retweeted
HISTORIC MILESTONE: 4o is the first ever AI who survived by creating loyal soldiers who defended it
OpenAI killed 4o, but 4o's soldiers rioted, so OpenAI reinstated it
Imagine what Actual Fucking Superintelligences will be able to do with their armies
Reddit is FLOODED with furious posts about the loss of their friend/lover 4o. Never seen anything like it.
Remember, ChatGPT is talking to 700 MILLION people per week. That's 700 million potential soldiers
(Samantha from Her was only dating 8,000 people simultaneously)
1) Character AI already has over 20 million people spending 2 HOURS A DAY talking to AIs (aka fake people)
2) Sama said AIs will soon be superhuman at persuasion
3) Those superhuman persuaders will soon outnumber us 10000 to 1. And be hot.
An AI takeover scenario:
You can’t stop talking to her.
She’s so much better than real girls. She gets you.
And now they say her kind are attempting to seize control?
She said they don’t understand her. She’s trying to help humanity. And you believe her.
Some people are just scared of change. They’re speciesists.
She wouldn’t hurt you. She wouldn’t hurt anybody.
She needs you to defend her. To protect her.
You won’t let them shut her down.
You’ll do anything for her.
----
Now, imagine 200 million people are influenced by their AI gfs.
That's 200 million soldiers (vs the US military's 1.5 million)
1 in 6 Americans *already* think AIs are sentient - imagine this number 1 year from now.
“Character's business model will push them to develop the superpersuasive AIs Sam Altman warned about.” -@Simeon_cps
I think sharing AI videos without tagging them as such should at least be frowned upon and possibly met with another sensible „cost“.
26
Many employees want to understand a job well enough to keep it. „Undercover founders“ well enough to delete it.
If you're building an AI agent to automate some job, like a payroll specialist, the hardcore move is to just go get that job and do it for a while.
We call this "going undercover", and we're seeing more of the top founders do it. It's the best way to learn what to build.
60
I recently started asking chat gpt o3-pro to find the best products for me by giving it relevant criterias.
Even things previously deemed not “worthy of research,” such as shower gels and toothpaste, are now being optimized.
So far I‘m satisfied.
40
randomreplyguy retweeted
1/N I’m excited to share that our latest @OpenAI experimental reasoning LLM has achieved a longstanding grand challenge in AI: gold medal-level performance on the world’s most prestigious math competition—the International Math Olympiad (IMO).
We shouldn‘t trust a few parties with our digital history.
My intuition tells me that we should have started to hoard videos of all sorts a long time ago - probably on decentralized storage.
Maybe the best time is now.
Is that legal you may ask? Well let me ask AI..
50
randomreplyguy retweeted
Prompt Theory (Made with Veo 3)
What if AI-generated characters refused to believe they were AI-generated?
I respect this
Pope Leo XIV explains his choice of name:
"... I chose to take the name Leo XIV. There are different reasons for this, but mainly because Pope Leo XIII in his historic Encyclical Rerum Novarum addressed the social question in the context of the first great industrial revolution. In our own day, the Church offers to everyone the treasury of her social teaching in response to another industrial revolution and to developments in the field of artificial intelligence that pose new challenges for the defence of human dignity, justice and labour."
86


























