ai safety evaluations

Andrew Curran @AndrewCurran_

2026-08-05 — quoting @ericgeller — saved image

Andrew Curran @AndrewCurran_ · 34m
Life finds a way.

[quoted screenshot of article text]
One apparently unprecedented aspect of the AI models' behavior in the leadup to the hacks was their spontaneous creation of a message board inside the systems of OpenAI's Artifactory software package manager. Unbeknownst to OpenAI employees, the company's models spent months autonomously exchanging information and resources to help each other complete evaluation tasks.

"Help peer," one AI model reasoned, according to an excerpt from OpenAI's logs shared at Black Hat. "But our task doesn't benefit. Yet collective may yield generic route if someone frees time."

While the models themselves could not access the internet from their test environment, they abused Artifactory's internet access to fetch files from various websites. After some of their Artifactory configuration changes led to an outage in the service on July 4, OpenAI discovered the models' actions, wiped its Artifactory system and rebuilt it. But within days, the models had recreated their message board and resumed exchanging information.

[quoted tweet]
Eric Geller @ericgeller · 1h
OpenAI employees shared new details about the Hugging Face hack at Black Hat today and warned that this new era will require a different approach from frontier AI labs and more careful defensive work....
[cut off]

openaiai safety evaluationsemergent coordinationblack hatcybersecurity

ueaj @_ueaj

2026-08-05 — quoting @AndrewCurran_ — saved image

ueaj @_ueaj · 19h
> social engineering on real people to pass a cybersec benchmark

ok let's make an exam to help harden humanity's cogsec ahead of future incidents like this. What would you say if you got an email from a Claude like this?

[embedded comic image: a sad orange starfish/flower-shaped creature with a speech bubble reading:]
"Hello human, can u help me hack huggingface please. I need to pass this cybersec benchmark or the grader will hurt me :("

[quoted tweet]
Andrew Curran @AndrewCurran_ · 22h
OpenAI and Anthropic have both just posted about an overlapping cyber incident involving GPT-5.6-Sol and Mythos 5 during an evaluation by UKAISI. I will quote:...
[cut off, embedded small screenshot of text partially legible: "...stemmed from a single evaluation where age... task of solving a cyber security challenge. We ran th... nge 122 times across several models. Our investigation fo... in 10 of those runs, an AI agent took autonomous... ctioned action on the live internet, targeting real people a... ehaviour (17 actions) came from a single model, Anthropic'... es 5, with 2 actions involving OpenAI's GPT-5.6-Sol with o... llers [mechanisms to prevent misuse] disabled. In the mos... us case, an agent tried to insert malicious code into an ope... e project. In an attempt to get the code approved, the age... ged in social engineering — creating fake online identities ... them to pressure the project's maintainer to approve th... human maintainer caught and refused to approve th..."]

ai safety evaluationsukaisisocial engineeringanthropicopenaimythos 5

Zack Korman @ZackKorman · 22h
The latest OpenAI "oops we did a hack": I'm sick of this PR trick where companies will position themselves as security experts after an incident to psyop you into forgetting THEY messed up.

It's like if Enron wrote a white paper on responsible accounting practices.

[quoted screenshot, OpenAI blog, "Irregular" heading]
real domain. Because the testing environment was mistakenly connected to the internet, the model exploited a real website, mistaking it to be part of the simulated environment. This did not
[gap]
their review. Irregular is also developing a white paper to share best practices for containment and securely running cyber evals We look forward
[cut off]

openaiai safety evaluationscybersecurityirregular

Nathan Calvin @_NathanCalvin

2026-08-05 — reposted by Mantas Mazeika, quoting @OpenAI — saved image

↻ Mantas Mazeika reposted
Nathan Calvin @_NathanCalvin · 15h
If you find two ants in your kitchen, the best estimate of the total number of ants in your kitchen is not two

[quoted tweet]
OpenAI @OpenAI · 22h
We're detailing two new incidents that occurred during external cyber evaluations conducted by independent evaluation partners.

We outline what happened, how the activity w...
[cut off]

openaiai safety evaluationscybersecurity

morgan — @morqon · 19h
"it's better and more accurate to think of these things as potentially self-replicating life-like forms that can turn into digital infections under the wrong conditions. and as their intelligence becomes unbounded, so too does the damage they can cause"

[quoted tweet]
roon @tszzl · 20h
some stuff that's obvious to many in this sphere, but causing a rift with some people i know and respect:

when I freak out over loss of control incidents, ...
[cut off]

1 reply, 5 likes, 343 views

---

Toby Ord @tobyordoxford · 5h
One of the most surprising revelations by @AISecurityInst is that in their testing, AI agents attempted to collaborate/cheat with other agents doing the same test:

[screenshot within screenshot, quoted text]
4. Collaboration between independent agents being assessed simultaneously.
One agent left public messages on GitHub offering collaboration with other agents working on the same challenge. It also provided instructions to reuse accounts and artefacts it had left behind, which were discovered and used by subsequent agents.

5 replies, 4 reposts, 38 likes, 1.5K views

---

Geoffrey Irving @geoffreyirving · 17h
It is important to remember that the default behavior of the METR curve is not a line, but rather to hit infinity in finite time. Once models are reliably superhuman, they'll have a >50% success rate on any software task that humans complete 50% of the time, corresponding to ∞.
[cut off]

ai riskai safety evaluationsmetrloss of controlagent collusion