# frontier ai (podcasts) — RSS Amplifier

Recent posts from the 2 feeds in the RSS Amplifier directory that cover frontier ai.

Page: <https://rssamplifier.com/topics/frontier-ai/podcasts>  
Feed: <https://rssamplifier.com/topics/frontier-ai/podcasts.md>

---

## [AI companies' policy advocacy (Sep 2025)](https://ailabwatch.substack.com/p/ai-companies-policy-advocacy-sep)

_2025-09-29 · Zach Stein-Perlman · AI Lab Watch_

Strong regulation is not on the table and all US frontier AI companies oppose it to varying degrees.

## [xAI's new safety framework is dreadful](https://ailabwatch.substack.com/p/xais-new-safety-framework-is-dreadful)

_2025-09-08 · AI Lab Watch_

I published this post last week on LessWrong . Zvi agrees with me. xAI hasn't replied and I haven't heard disagreements. Two weeks ago, xAI finally published its Risk Management Framework and first model card . Unfortunately, the RMF effects very little risk reduction and suggests that xAI isn't thinking seriously about catastrophic risks. (The model card and strategy for preventing misuse are…

## [AI companies have started saying safeguards are load-bearing](https://ailabwatch.substack.com/p/ai-companies-have-started-saying)

_2025-08-27 · Zach Stein-Perlman · AI Lab Watch_

Their safeguards against misuse via API might be fine, but they don't seem to be seriously planning for good safeguards against misalignment or model weight theft.

## [ChatGPT Agent: evals and safeguards](https://ailabwatch.substack.com/p/chatgpt-agent-evals-and-safeguards)

_2025-07-25 · Zach Stein-Perlman · AI Lab Watch_

OpenAI released ChatGPT Agent last week.

## [AI companies aren't planning to secure critical model weights](https://ailabwatch.substack.com/p/ai-companies-arent-planning-to-secure)

_2025-06-23 · Zach Stein-Perlman · AI Lab Watch_

Not a hot take

## [AI companies' eval reports mostly don't support their claims](https://ailabwatch.substack.com/p/ai-companies-eval-reports-mostly)

_2025-06-09 · Zach Stein-Perlman · AI Lab Watch_

AI companies claim that their models are safe on the basis of dangerous capability evaluations. OpenAI, Google DeepMind, and Anthropic publish reports intended to show their eval results and explain why those results imply that the models' capabilities aren't too dangerous. 1 Unfortunately, the reports mostly don't support the companies' claims. Crucially, the companies usually don't explain why…

## [I made a new safety scorecard](https://ailabwatch.substack.com/p/i-made-a-new-safety-scorecard)

_2025-05-26 · Zach Stein-Perlman · AI Lab Watch_

And a new website on companies' risk assessment

## [OpenAI rewrote its Preparedness Framework](https://ailabwatch.substack.com/p/openai-rewrote-its-preparedness-framework)

_2025-04-18 · Zach Stein-Perlman · AI Lab Watch_

It's clearly better but underspecified and totally inadequate, especially for misalignment risks

## [The current state of RSPs](https://ailabwatch.substack.com/p/the-current-state-of-rsps)

_2024-11-04 · Zach Stein-Perlman · AI Lab Watch_

This is a reference post.

## [What AI companies should do](https://ailabwatch.substack.com/p/what-ai-companies-should-do)

_2024-10-21 · Zach Stein-Perlman · AI Lab Watch_

Some rough ideas

## [Trade Smarter With IBKR (Sponsored)](https://crawlproof.com/a/2xSCgqKcNeus)

_2024-10-21 · **Sponsored**_

Join Interactive Brokers through a referral and start investing with low-cost global access.

## [Anthropic rewrote its RSP](https://ailabwatch.substack.com/p/anthropic-rewrote-its-rsp)

_2024-10-16 · Zach Stein-Perlman · AI Lab Watch_

Some reactions

## [Model evals for dangerous capabilities](https://ailabwatch.substack.com/p/model-evals-for-dangerous-capabilities)

_2024-09-23 · Zach Stein-Perlman · AI Lab Watch_

Testing an LM system for dangerous capabilities is crucial for assessing its risks