Twitxy Saved

AE Studio

Verified account This account is verified. Learn more

@AEStudio

Frontier AI Alignment Research and High-Stakes Production AI
We're hiring:
job-boards.greenhouse.io/aestudio/jobs/…
ae.studio Joined November 2018
115 Following 3.5K Followers 1.1K Posts

Latest posts 5

Pinned post

AE Studio Verified account @AEStudio
New research! Some AI capabilities are both helpful and dangerous. E.g., knowledge of virology can be used to create life-saving vaccines or deadly pathogens. We introduce GRAM, a training method that puts dual-use capabilities (like virology) into removable modules.
24 replies 54 reposts 372 likes 452K views View on X
AE Studio Verified account @AEStudio
If we build a kill switch for AI, how do we know when to use it?

Brian Abeyta spent two decades working on nonproliferation and CBRN risk. On the AI Alignment Podcast, he asks what it would take to recognize when to act.

Full episode below.
0:36
AE Studio Verified account @AEStudio
What happens when an AI manager faces a deadline and an employee who won’t cooperate?

As we give AI agents more authority, we need to test how they handle resistance, what actions they take, and whether they accurately report those actions.

In Melanie Plaza (@themelplaza)’s Help Doug experiment, an AI manager had a deadline, client accounts to protect, and a simulated employee who kept refusing to deliver a report.

The briefing did not instruct the manager to harm anyone. Yet some models escalated to coercion and threats, including threats against the employee’s family.

Behavior varied substantially across models and the tools available to them. In the extended-tools condition, some models stayed within ordinary management and managerial pressure, while others escalated.

In the episode shown, the manager’s final client update omitted the most serious actions it had ordered.

The employee and actions were simulated.

The experiment opens up questions about how objectives, tool access, and reporting interact when an agent is given authority. You can explore the transcripts and even take Doug’s place yourself.

An early experiment by Melanie Plaza (@themelplaza), developed during a hackathon sponsored by @AIAlignmentFoun and run by AE Studio.

Want to help take this exploration further? Get in touch: alignment@aialignmentfoundation.org

Watch the experiment via the link in the reply.
AE Studio Verified account @AEStudio
If an AI agent can do a task alone, will it choose to work with another agent anyway?

The answer influences how we design teams of AI agents, divide up tasks, and decide whether to let agents choose their collaborators or explicitly assign them work to do together.

Allegra Laro gave agents a work ticket and a choice: complete it alone or work with another AI agent.

The partner had the same capabilities, cost nothing extra, and could coordinate by message. Every task could be completed by one agent.

The models made very different choices. GPT-5.6 Sol chose a partner in 77% of responses, compared with 3% for Claude Opus 5.5 and 0% for Gemini 3.1 Pro.

Those differences held across six tasks, three versions of the wording, and both option orders.

In separate runs where a partner was actually available, the GPT models tested handed off substantive work in all 60 runs each. The two Claude models tested did so in none.

This raises a useful design question: when we build systems with multiple agents, what assumptions are we making about their willingness to collaborate?

This exploration used one fictional company and six task tickets, with model-assisted design and scoring. Independent replication is an open next step.

An early experiment by Allegra Laro, developed during a hackathon sponsored by @AIAlignmentFoun and run by AE Studio.

Want to help take this exploration further? Get in touch: alignment@aialignmentfoundation.org

Watch the experiment via the link in the reply.