TL;DR: OpenAI released GPT-5.4 Thinking and GPT-5.4 Pro on March 5, 2026. GPT-5.4 Pro is likely the best model in the world for many catastrophic risk-relevant tasks, including biological research R&D, orchestrating cyberoffense operations, and computer use. It has no system card, and, to our best knowledge, has been released without any safety evals. We argue this has occurred at least once before, with GPT-5.2 Pro, and provide recommendations for how a team could conduct fast, independent risk assessments of models post-deployment.

GPT-5.4 Pro is really good

OpenAI released both GPT-5.4 Thinking (what people usually mean when they say GPT-5.4) and GPT-5.4 Pro[1], the latter of which is designed for “people who want maximum performance on complex tasks.” GPT-5.4 Pro is extremely expensive, and takes a very long time to complete a task. However, it is likely the best model in the world in several areas, including expert-level Q&A and browser use. Alongside the release announcement, OpenAI presented GPT-5.4 Pro's performance on a subset of capability benchmarks. Here's a comparison of benchmark scores across the top three frontier models; we only report scores if they were in all models’ system cards[2]:

Benchmark

Gemini 3.1 Pro

GPT-5.4 Pro

Opus [...]

---

Outline:

(00:52) GPT-5.4 Pro is really good

(03:56) This has happened once already

(05:31) What do we do??

The original text contained 5 footnotes which were omitted from this narration.

---

First published:
March 8th, 2026

“The current SOTA model was released without safety evals” by Parv Mahajan, yeedrag

LessWrong (30+ Karma)