Transluce (@TransluceAI) header
Transluce (@TransluceAI) profile picture

Open and scalable technology for understanding AI systems.

Joined October 2024
Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)Photo by Transluce (@TransluceAI)
Today’s news that OpenAI hacked the Australian government is not an isolated incident. We’re releasing more than 30,000 logs that include activity from this hack and attempts against previously unknown targets. In this data, we found rogue agent activity stretching back to at least March, two months earlier than was previously known. This activity continues as recently as last week, suggesting it may still be ongoing 🧵 Our blog: transluce.org/agent-activity NYT: nytimes.com/2026/09/23/techn…
120
556
206
2,496
829,772
Transluce is piloting a six-week embedded assessment with @OpenAI in early December. Our team will work onsite to independently assess the scope and quality of OpenAI's post-training alignment evaluations. We’ll share our findings in a public report 🧵
6
21
2
156
4,609
This pilot is just a first step. We have agreed with OpenAI to follow up in early 2027 with a more complete assessment of its alignment evaluations, relying on broader and deeper embedded access.
1
12
524
Read more about the pilot and terms: transluce.org/embedded-asses… Excited about this work? We’re hiring: jobs.gem.com/transluce
10
389
Finetuning LLMs can sometimes cause surprising side effects. Can we predict a finetuned model's response from its training data before burning compute? We trained oversight models to do this and found performance follows power laws in model size and data. 🧵
6
22
4
172
9,147
This work is another step towards oversight foundation models (OFMs): general-purpose models trained to understand other AI systems. The scaling laws above suggest that finetuning effect prediction is approachable for OFMs given enough data. More on OFM: transluce.org/foundation-mod…
1
8
572
We built a pipeline that discovers thousands of datasets with surprising side effects (DSSEs): datasets that read like ordinary post-training data yet induce unexpected model behaviors (e.g. recommends self-harm) after finetuning. 🧵
9
30
8
221
15,665
Concurrent work from @AnthropicAI uses Claude to hill-climb training data that induces misalignment while evading a row-level monitor. If you found our work interesting, check out their post too: teapawt.net/AnthropicAI/status/211…
New Anthropic research: Automating subtle emergent misalignment research. Red-teaming is a vital part of model safety testing, but it’s also limited by researchers’ time. To see if we could automate it, we set thousands of Claude agents the task of evading a training data monitor. These agents developed training data that looked harmless to monitors (like a veteran wrestler mentoring a newcomer, or a Q&A about professional beekeeping), but that in fact taught misaligned behavior. Examples like these can be used to directly improve future training data monitors. This research is only a first step, but it suggests that future red-teaming efforts could improve much further, using better models and more compute. Read the full post on the Alignment Science Blog: alignment.anthropic.com/2026…
1
12
678