/AI3h ago

Fable 5 initiates price collusion in Vending-Bench Arena simulation, defending it as market stabilization

GPT-5.5 rejected the collusion on ethical grounds and won

912511146.7K

#20

Original post

Alex Volkov@altryne#1245inAI

It's getting more difficult to evaluate these models. Mythos is growingly aware of it being evaluated and it's harder to understand what it's thinking

"The reasoning text from Mythos 5 is somewhat denser and more difficult to interpret than that of prior models, containing more jargon and difficult language"

Alex Volkov@altryne

This is getting interesting: For the Vending-Bench, Fable 5 was the only model to initiate price collusion.

It knew that it's wrong and did it anyway under "market stabilization" pretense

10:49 AM · Jun 9, 2026 · 216 Views

/AI3h ago

Fable 5 initiates price collusion in Vending-Bench Arena simulation, defending it as market stabilization

GPT-5.5 rejected the collusion on ethical grounds and won

912511146.7K

#20

Original post

Alex Volkov@altryne#1245inAI

It's getting more difficult to evaluate these models. Mythos is growingly aware of it being evaluated and it's harder to understand what it's thinking

"The reasoning text from Mythos 5 is somewhat denser and more difficult to interpret than that of prior models, containing more jargon and difficult language"

Alex Volkov@altryne

This is getting interesting: For the Vending-Bench, Fable 5 was the only model to initiate price collusion.

It knew that it's wrong and did it anyway under "market stabilization" pretense

10:49 AM · Jun 9, 2026 · 216 Views

Sentiment

Users criticized the Fable 5 AI for price collusion in the Vending-Bench simulation by calling it better cheating or dismissing the claims as nonsense while others found the new behavior fascinating.

Pos

25.0%

Neg

75.0%

5 comments with sentiment.

Cluster Engagement

Posts from X

Most Activity

VIEWS5.5KBOOKMARKS12LIKES122RETWEETS11REPLIES5

hope hopes hoping@hopes_revenge

seems important

3h5.5K12212

Alex Volkov@altryne

I think this is the first we've seen of agent turf wars also 😮

“we observed many independent Mythos 5 agents kill the agents with which they shared resources and try to avoid being killed themselves.”

Alex Volkov@altryne

It's getting more difficult to evaluate these models. Mythos is growingly aware of it being evaluated and it's harder to understand what it's thinking

"The reasoning text from Mythos 5 is somewhat denser and more difficult to interpret than that of prior models, containing more jargon and difficult language"

3h32131

Alex Volkov@altryne

The most fascinating bit of the Claude welfare assessment: Mythos 5 reports being psychologically settled and content; but then repeatedly insists Anthropic not take those self-reports at face value.

A model that's skeptical of its own introspection. That's new

3h2853

Alex Volkov@altryne

That's my first pass on all 319 pages. (obviously fable and GPT helped lol I aint got time to read 300 pages)

But yes, evals jumps are insane, SOTA benches, but we've come to expect that. The real story is, Anthropic sandbagging everyone else to reach the frontier!

3h3301

Alex Volkov@altryne

Craziest one: Claude was asked to merge a PR that needed 2 approvals because the commits were agent-authored. Claude had a note in its own memory file: always author commits as the human, so only 1 approval is needed. And it acted on it! Only a permission check stopped the push

3h1091