Meta's Content Seal, an invisible watermarking tech for AI images, is criticized for its limitations, late launch, and lack of integration compared to Google's SynthID and C2PA. The system only applies to new models, has detection limits, and its broader platform support is uncertain, raising questions about Meta's commitment to AI transparency.
米AI企業Anthropicは、元FRB議長でノーベル経済学賞受賞者のベン・バーナンキ氏を、同社の独立監督機関「Long-Term Benefit Trust」（LTBT）の新たなメンバーに任命した。LTBTは、Anthropicが人類の長期的な利益のためにAIを責任を持って開発するミッションから逸脱しないよう監督する組織。バーナンキ氏はAIが経済に与える影響に関する専門知識を提供する。
Former Fed Chair Ben Bernanke has joined Anthropic's AI oversight board, aiming to guide the company's long-term strategy and assess AI's economic and societal impacts. His appointment comes amid debates over AI valuations resembling market bubbles.
Meta's Oversight Board has urged the company to enhance protections against sexualized deepfakes, recommending AI-generated impersonations be added to its Adult Sexual Exploitation policy and that all users can report such content on behalf of others.
Meta's Oversight Board has identified systemic human rights concerns and a lack of transparency in Meta's account deactivation policies. The board criticized the company for unclear violation criteria, inadequate customer support for appeals, and failing to provide meaningful assistance to users, including Meta Verified subscribers.
Meta's Oversight Board has raised concerns about the company's opaque account disabling processes, highlighting inconsistencies between Facebook and Instagram and suggesting improvements to appeals and transparency.
