Page 12: Research news on Generative AI misinformation

Generative AI misinformation concerns the creation and dissemination of synthetic media—such as hyperrealistic deepfake images, videos, and voices—used to deceive, manipulate, or exploit individuals and publics. Work in this area examines AI-enabled impersonation, political and commercial disinformation, scams, and non-consensual explicit content, as well as their psychological and societal impacts. It also investigates technical and sociotechnical defenses, including detection models, watermarking and provenance systems, security against data poisoning and backdoors, and human-centered interventions to preserve trust and information integrity.

Internet

Grok spews misinformation about deadly Australia shooting

Elon Musk's AI chatbot Grok churned out misinformation about Australia's Bondi Beach mass shooting, misidentifying a key figure who saved lives and falsely claiming that a victim staged his injuries, researchers said Tuesday.

Internet

Instagram users given new algorithm controls

Instagram on Wednesday unveiled a new AI-powered feature that lets users view and adjust the algorithm shaping their Reels feed, calling it a pioneering move toward greater user control.

Consumer & Gadgets

How 'everyday AI' encourages overconsumption

From automatically generated overviews to chatbots in spreadsheets, so-called artificial intelligence is increasingly being integrated into our watches, phones, home assistants and other smart devices.

Consumer & Gadgets

Using food to uncover AI's cultural blind spots

CISPA researcher Tejumade Àfọ̀njá has co-authored a new international study that uses food as a starting point to reveal significant cultural blind spots in today's AI systems. The study also introduces a new participatory ...

page 12 from 26