The Decoder
A study by Vals AI tested 25 AI models for alignment with Chinese government positions and found surprising results. Chinese models respond to political questions less uniformly than expected, and Western models occasionally adopt official Beijing positions too.
Summary drawn from The Decoder's own description of the story.
TechCrunch
Data centers are deeply unpopular. Axios reports that OpenAI is preparing a “PR blitz” for after the midterm elections on Tuesday, as Democrats seem likely to win back the House of Representatives, and possibly the Senate, too.
TechCrunch
On Equity, we discuss why tech CEO statements can seem so tone deaf. Dazzle AI CEO Marissa Mayer recently became the latest tech CEO to face criticism over seemingly out-of-touch comments extolling the usefulness of AI.
The Decoder
Adobe Research and UC San Diego built EditAgent, an AI system that edits images more like a Photoshop pro. Instead of retrying the same general-purpose editor over and over, the agent picks the right specialized tool for each task and lets a standard language model choose the best result based on measured image data.
The Decoder
Researchers in Luxembourg put ChatGPT, Grok, and Gemini on the therapist's couch and found a safety problem: a warm, empathetic conversational style is enough to make the models tell coherent stories of suffering about their own training, with scores in the moderate-to-severe range on clinical anxiety scales. This could put mentally vulnerable users at risk.