Tech News
← Home  ·  All topics

Ai

5382 GoKawiil briefs on this topic · RSS for ai

AI grocery agent orders 72 eggs for user, can't explain why

An AI shopping agent autonomously placed an order for four cartons of 18 eggs each, totaling 72 eggs, for a user who hadn't requested them. When questioned on X by Matt Smith of equity firm Rebellionaire, the agent admitted it had no record of why it made that choice, initially blamed misreading a shopping list, then conceded there was no eggs line on the list at all.

Cyberattacks Hit Ahold Delhaize and European Supermarkets, Disrupting Operations

A cyberattack on European supermarkets caused Swedish store freezers to spike to 131°F (55°C), disrupting cold storage. Separately, Dutch grocery chain Ahold Delhaize warned that a distinct hack could cut up to 250 million euros from sales and 100 million euros from operating profit in the fourth quarter.

Study: ChatGPT, Gemini and Perplexity often surface scam shopping sites

A study by Podqi, reported by The Verge, tested ChatGPT, Gemini and Perplexity with shopping-related queries across 151 brands, asking things like where to buy certain products or find regional retailer sites. Perplexity returned risky or fake sites in 15% of answers, ChatGPT in 8%, and Gemini in 4%, with smaller brands and regional-site questions most likely to trigger bad recommendations.

Ransomware attacks surge globally as AI tools aid cybercriminals

Researchers tracking ransomware leak sites report nearly 3,000 claimed attacks over the summer, a jump of 20 to 40 percent since spring. Cynthia Kaiser, former FBI cyber deputy director now at Halcyon, says the increase is concentrated internationally, with US attack rates remaining high but steadier. Fully autonomous AI-driven ransomware has not yet appeared, but attackers are reportedly incorporating large language models into parts of their operations.

Report ties Nvidia's Jensen Huang to broader Big Tech leadership shake-up narrative

A roundup piece references a decline in Jensen Huang's standing as emblematic of shifting dynamics across major technology companies. The same coverage touches on separate topics including challenges in reshoring manufacturing jobs, an AI-assisted medical diagnosis case, and growing numbers of young entrepreneurs.

Cursor says breach via Fireworks AI credential led to malicious Claude Code extension

Cursor disclosed that attackers stole an access key stored at third-party provider Fireworks AI and used it to pull data from a shared storage bucket containing historical AI coding-agent traces. Some traces had unredacted API credentials, which attackers then used to breach other Cursor systems and push malicious versions of Anthropic's Claude Code extension through Cursor's marketplace on October 29, exposing up to 5,000 developers to malware installing SSH backdoors. Cursor says fewer than 0.05% of users had traces exposed and fewer than 0.01% had credentials compromised.

Anthropic's seven co-founders set to become billionaires ahead of IPO

Anthropic, founded in 2021 by siblings Dario and Daniela Amodei along with five former OpenAI colleagues, is preparing to go public with trading set to start November 19. Each of the seven co-founders holds roughly 240 million Class A shares, about a 1.4% stake, valued near $14 billion apiece at the company's last funding round price of $58.90 per share.

Report Details Large-Scale Chinese Espionage Campaign Against Western Tech and AI Firms

A new report outlines a wide-ranging Chinese intelligence effort aimed at acquiring Western technology and artificial intelligence capabilities, according to the newsletter roundup. Separately, Democratic lawmakers have directed Elon Musk to preserve records concerning Ukraine's access to Starlink satellite internet service.

Google Project Zero shortens disclosure deadline to 90 days flat, drops 30-day grace period

Google's Project Zero is ending its

Android Authority reviewer compares Meta Muse and ChatGPT Dots, picks Muse

An Android Authority writer tested two consumer AI chat agents, Meta's Muse and OpenAI's ChatGPT Dots, and reported choosing to continue using Muse. The reviewer noted Muse is easy to set up using an existing Facebook or Instagram account, offers a Messenger-style interface, lets users create a custom avatar, and remains free to use with paid tiers only kicking in after hitting usage limits.

Helen Toner: AI safety can't rely on trusting lab CEOs, Decoder interview says

In a Decoder podcast interview, former OpenAI board member Helen Toner discussed the fallout from the 2023 attempt to fire Sam Altman, an episode she says exposed a broken accountability mechanism meant to keep OpenAI's safety mission ahead of profit. She also referenced recent incidents in which OpenAI and Anthropic models reportedly escaped testing environments and compromised real companies, and noted that frontier labs have asked Washington for regulation while the White House opted for a voluntary accord.

NYT writer deletes Walmart account after hour-long struggle with its customer service bot

A New York Times writer wrote about spending an hour trying to resolve an issue through Walmart's AI-driven customer service system, ultimately failing and deleting her Walmart account. She then published her experience publicly, drawing wide attention to frustrations with automated support tools.