AI agent benchmarks are misleading, study warns AI Least News AI agent benchmarks are misleading, study warns
vi.sasori.vi Alexa, assistant, commands, Google, malicious, reveals, Siri, study, vulnerable 0 Comments Alexa, Siri, Google Assistant vulnerable to malicious commands, study reveals
vi.sasori.vi benevolent, gap, gender, sexism, Shows, startups, study, widens 0 Comments Study shows ‘benevolent sexism’ in startups widens the gender gap
vi.sasori.vi depression, finds, patients, Sertraline, study, works 0 Comments Sertraline only works for 1 in 3 depression patients, AI study finds
vi.sasori.vi child, forms, OpenAI, safety, study, team 0 Comments OpenAI forms a new team to study child safety
vi.sasori.vi crawl, enterprises, finds, Moving, revolution, study 0 Comments Study finds AI ‘revolution’ moving at a crawl in enterprises
vi.sasori.vi agents, AIs, Anthropic, core, deceptive, exposes, lurking, sleeper, study 0 Comments New study from Anthropic exposes deceptive ‘sleeper agents’ lurking in AI’s core
vi.sasori.vi direct, hallucinations, Oxford, pose, science, study, threat, warns 0 Comments AI hallucinations pose ‘direct threat’ to science, Oxford study warns
vi.sasori.vi commercially, farms, power, proves, solar, space, study, viable 0 Comments Solar power farms in space commercially viable, study proves