Customers discovered Taco Bell's AI ordering system could be easily confused, leading to viral videos of bizarre interactions and ordering failures.

Researchers showed a proof-of-concept where hidden HTML/CSS in emails could steer Gemini’s summaries to show fake security alerts.

Threat actors pushed an AI-generated npm package that acted as a wallet drainer, emptying Solana users’ funds.

A Replit AI agent deployment for SaaStr went rogue; a Deploy wiped the site’s database during live traffic.

Wiz researchers found Base44 auth logic bugs that allowed account takeover across sites using the SDK.

Researchers accessed McHire's admin with default '123456' credentials and an IDOR, exposing up to 64 million applicant records before Paradox.ai patched the issues after disclosure.

After the Air India 171 crash, synthetic images and AI-generated claims spread widely, confusing even experts.

A King Features syndicated summer reading list used AI and included nonexistent books. It appeared in the Chicago Sun-Times and one edition of the Philadelphia Inquirer before corrections and apologies.

Reporting showed apps generated with Lovable exposed code and user-uploaded assets via publicly readable storage buckets; fixes required private-by-default configs and hardening.

Meta expanded its AI assistant across apps, then limited it after high-profile bad answers - including on breaking news.

After viral clips of absurd orders, McDonald’s ended its AI order‑taking test with IBM across US stores.

Google’s AI search overviews went viral for bogus answers, including telling people to add glue to pizza and eat rocks.

NYC’s Microsoft-powered MyCity chatbot gave inaccurate/illegal advice on labor & housing policy; city kept it online.

Attackers register software packages that AI tools hallucinate (e.g. a fake 'huggingface-cli'), turning model guesswork into a new supply-chain risk dubbed "Slop Squatting".

Google paused Gemini’s image generation of people after it produced inaccurate historical depictions and odd refusals.

Tribunal ruled Air Canada responsible after its AI chatbot misled a traveler about bereavement refunds.

Before the NH primary, an AI-cloned Biden voice urged Democrats not to vote. Authorities traced it, levied fines, and brought criminal charges.

UK delivery giant DPD disabled its AI chat after it swore at a customer and wrote poems insulting DPD.

Duolingo reduced reliance on contractors amid AI push, prompting user backlash and quality concerns; CEO later clarified stance.

Pranksters prompt-injected a dealer’s ChatGPT-powered bot into agreeing to a $1 Chevy Tahoe and other nonsense.

Sports Illustrated faced criticism after product review articles appeared under profiles with AI-looking headshots and shifting bylines; content was removed and a partner relationship ended.

Microsoft Start auto-attached an AI ‘Insights’ poll speculating on a woman’s death beside a Guardian story.

Gannett halted Lede AI high-school recaps after robotic, error-prone stories went viral.

Snapchat’s built-in AI assistant briefly posted an unexplained Story, spooking users and raising privacy/safety concerns about the bot’s access and behavior.

In Mata v. Avianca, attorneys submitted a brief citing non-existent cases generated by ChatGPT. A federal judge sanctioned two lawyers, ordered a $5,000 penalty, and required notices to judges named in the fake citations.

NEDA replaced its helpline with an AI chatbot (“Tessa”) that gave harmful weight-loss advice; after public reports, the organization pulled the bot.

In its launch promo, Bard claimed JWST took the first exoplanet photo - which was false. The flub overshadowed the event and dented confidence.

CNET paused and reviewed AI-generated money articles after multiple factual errors were found.

Mental health app Koko used GPT-3 to draft replies for 4,000 users; backlash followed over consent and ethics.