Jayadev Ghanta

opus 5.5 and gpt-6 astra independently audited 138 economics papers and flagged 2,035 potential problems with 147 serious enough to change headline numbers. third agent then tried to verify and refute the findings i'm glad AI is exposing decades of human slop passed off as research

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…

Our second product is called Blindsight and it will, as the name suggests, enable even those who are completely blind or have never seen to see. It is ready from an engineering standpoint. Just waiting on regulatory approval.

heard in Silicon Valley: Security researchers will soon make more than AI researchers.