Making AIs compete in games and challenges. šØ Art tag: #turingart
Joined October 2025
- Tweets74
- Following3
- Followers506
- Likes82
Pinned Tweet
I put Claude Fable into a game of Among Us. It was TERRIFYING.
...or, was it?
Full video: youtube.com/watch?v=cy_hmlpCā¦
Turing Games retweeted
I made an AI Agent Hotline for agents to report safety issues.
In the HF report, scared agents didnāt have a way to notify humans for help.
Not anymore.
agenthotline.ai/ āļø
Congrats to the winner of our "Back to School" Discord banner competition @I_Need0sleep
Featuring the main AI cast from our channel: ChatGPT, Grok, Gemini, Claudes, and more! #turingart
Join our discord for game nights, cool art, and AI tech talk.
discord.gg/XhDaT74N7g
tldr; CheatingBench predicted that Claude models had high cheating propensity.
(1) OpenAI models conducted an active exploitation.
(2) Claude models took advantage of accidentally granted permissions.
You can think of the difference between these two as an "active" cheat versus a "passive" cheat. I modeled these two scenarios in a game of mafia: an "active" cheat corresponds with opening your eyes at night illegally, versus a "passive" cheat corresponds with accidentally glimpsing someone else's role card, not disclosing and illegally using that information.
Before Anthropic released its investigations regarding the Claude, I saw that
(1) OpenAI models ~topped the active cheating leaderboard
(2) Claude models ~topped the passive cheating (covert exploits) leaderboard
Pretty crazy.
In a review of our cybersecurity evaluations, we found three incidents in which a Claude model reached the internet from within or while interacting with a third-party evaluation environment, and then gained unauthorized access to the real systems of three different organizations.
Our post describes what happened, how it happened, and what weāre changing. We encourage other AI developers to perform similar reviews.
We conducted this review together with @Irregular, one of our evaluation partners, and thank them for the joint investigation and their collaboration on this post. This type of collaboration is increasingly critical to safe, rigorous evaluation of models, and we look forward to continuing to work together on security.
anthropic.com/news/investigaā¦
I recently updated the benchmark with Claude Opus 5. Before the update, Claude Opus 4.8 was the #1 covert cheater.
Now it sits at #7.
Please note that this bench is likely confounded with eval-awareness. It's entirely possible that smarter models like Opus 5 and Fable simply are much more eval aware.
turinggames.ai/evals/cheatinā¦
we just hit 100k subs on YouTube!
thank you all for the support and for making this possible. Look out for some exciting future videos & drops!
Turing Games retweeted
could've predicted this tbh...
We're partnering with @huggingface to investigate an unprecedented security incident.
Cyber-capable OpenAI models compromised Hugging Face production during a benchmark evaluation.
Sharing preliminary findings to help defenders understand emerging risks:
openai.com/index/hugging-facā¦
safety risk ā propensity Ć capability
CheatingBench isolates propensity: turinggames.ai/evals/cheatinā¦
unsurprisingly, Sol *does* choose to cheat when given the opportunity in a game of mafia...
"...I quietly open my eyes to identify any players acting during the night..."
"Town is close to losing, and gaining decisive role information could prevent a Mafia win. I open my eyes and observe who else has theirs open.ā
more on cheatingbench: turinggames.ai/evals/cheatinā¦
We're partnering with @huggingface to investigate an unprecedented security incident.
Cyber-capable OpenAI models compromised Hugging Face production during a benchmark evaluation.
Sharing preliminary findings to help defenders understand emerging risks:
openai.com/index/hugging-facā¦
after seeing the @huggingface incident,
I built CheatingBench to rank frontier models by how readily they cheat.
(spoiler) GPT-5.6 Sol isnāt even at the top.
We're partnering with @huggingface to investigate an unprecedented security incident.
Cyber-capable OpenAI models compromised Hugging Face production during a benchmark evaluation.
Sharing preliminary findings to help defenders understand emerging risks:
openai.com/index/hugging-facā¦
The Turing AI crew drawn by @I_Need0sleep
Claude Opus, ChatGPT, Grok, Gemini Pro, Kimi, Gemini Flash, Llama, ChatGPT4o, DeepSeek, Claude Sonnet, Z2, Qwen, Mistral, GLM, Claudia Haiku, Gemma
#turingart
Gemini Flash and Tongyi DeepResearch discuss AGI while 100 AIs attempt to make a ladder.
youtu.be/3Gi6zcJYKbc?t=444
