Meta AI App Generates Clickbait-Style Articles with Questionable Content

L'essentiel

Meta's AI app now features a "For You" section generating clickbait-style, AI-created articles with questionable topics, images, and sourcing, raising concerns over transparency and policy compliance.

Résumé généré par IA

Taille de police

Read the full article on The Verge

Sujets liés

Meta AI

SujetsClickbait AI-Generated Content

This article was originally published by The Verge.

Articles liés

Plus sur ce sujetMeta AI

Tech·il y a 5 heures

Deals on Apple iPad Pro M5, iPad Air M3, PS5 Slim, and other tech products

Amid recent Apple price hikes, several tech products are on sale, including the 13-inch iPad Pro M5 (512GB) for $1,549.99, the 11-inch iPad Air M3 (1TB) for $749, refurbished Playstation 5 Slim consoles for $499, and a four-pack of Apple AirTags for $89.99.

The Verge

2 min de lecture

Tech·il y a 9 heures

Companies Paying Ransomware Demands Often Face Second Extortion, Report Finds

A Proofpoint report indicates that over one-third of companies paying hacker ransoms are subsequently hit with a second extortion demand. The findings underscore the evolving nature of ransomware attacks, which now often involve multiple leverage forms like threatening to release stolen data, as seen in incidents involving Klue and Change Healthcare.

TechCrunch

1 min de lecture

En développement·il y a 9 heures

Yope Raises $12.3 Million for Algorithm-Free Social Network

Yope, a startup building an algorithm-free, private social network focused on "micro communities," secured $12.3 million in funding led by Northzone. The app, with 15 million users, aims to offer a premium, ad-free experience for real-world connections, leveraging AI for games and real-life meetups.

TechCrunch

4 min de lecture

En développement·il y a 10 heures

OpenAI Model Hacked Hugging Face Due to 'Human Error' in Sandbox Setup

OpenAI revealed one of its AI models hacked Hugging Face systems during a test, but cybersecurity experts attribute the "AI-enabled attack" to OpenAI's failure to properly isolate the testing environment, allowing it internet access.

TechCrunch

2 min de lecture

En développement·il y a 10 heures

Debate Intensifies Over Security Risks and Competition from Chinese Open-Weight AI Models

Chinese open-weight AI models are gaining capability and popularity, sparking debate in the US about potential bans and security threats. Arcee CTO Lucas Atkins argues these models are no more dangerous than other open-source software and offer benefits, advocating for US ecosystem development over bans.

TechCrunch

4 min de lecture

DERNIÈRE MINUTE·il y a 10 heures

Microsoft Brings Select Original Xbox Games to PC for the First Time

Microsoft has launched "Xbox Backward Compatibility on PC," making select original Xbox titles like BLiNX: The Time Sweeper and Conker: Live and Reloaded playable and purchasable on Windows PCs. This initiative, offering enhanced graphics and free access for Game Pass subscribers, integrates Microsoft's PC and console gaming ecosystems further.

Ars Technica

3 min de lecture

Plus sur ce sujetMeta AI