OpenAI flags concerning new AI behavior and vows to track it more closely

Mint - Sep 17, 2026

More News

Sep 17, 2026

OpenAI discloses 6 reports of AI models' 'unexpected or concerning' behavior

OpenAI published six new reports of artificial intelligence models showing “unexpected or concerning” behavior Wednesday as pressure grows on AI firms to be more transparent about the development process. The ChatGPT maker disclosed the reports as part of its new framework for tracking and disclosing instances of model misalignment, which occurs when an AI system...

Sep 17, 2026

On ‘Cupertino,’ Creators of the ‘The Good Wife’ Go to Silicon Valley

The prolific producers Michelle and Robert King target tech bigwigs in their newest show. “They’re really the stupidest and dumbest people,” Robert King said.

Sep 17, 2026

Judge orders Google to change ad auction rules

A federal judge has ordered Google to change its rules for auctioning online advertising on its platforms after finding last year that the search giant had an illegal monopoly over the technology connecting publishers and advertisers. The remedies were made public late Wednesday, when U.S. District Judge Leonie Brinkema unsealed her opinion from earlier this...

Sep 17, 2026

King Charles warns of 'existential danger' of AI falling into wrong hands

The comments were made at a summit in Ayrshire with participants from AI giants including Nvidia, OpenAI and Anthropic.

Sep 17, 2026

4 Strategies to Battle ‘Techflation’

With phone and computer prices soaring, it helps to know the true cost of owning electronics so you can bring those numbers down.

Sep 17, 2026

SEC unveils exemption for tokenized stock trading

The Securities and Exchange Commission (SEC) unveiled its long-awaited “innovation exemption” Thursday, a new order aimed at clearing the way for tokenized stock trading in the U.S. The order creates a pair of five-year exemptions. The first exempts platforms that host tokenized stock trading from the commission’s definition of an exchange, while the second exempts...

Sep 17, 2026

TUI Airline and GO7 launch industry-fi rst airline-controlled connectivity partnership

TUI Airline, part of one of the world’s leading travel groups, and GO7, the growth platform for airlines, today announced an industry fi rst: TUI Airline is the fi rst carrier to go live with GO7’s airline-controlled connection model. The partnership enables customers to combine two fl ights in one smoother journey, including through-checked baggage, while allowing airlines to retain ownership of pricing, settlement, operations and the customer relationship.

Sep 17, 2026

Critical Unbound DNSSEC Validator Flaw Could Allow RCE via a Malicious DNS Zone

Every release of the Unbound DNS resolver before 1.26.1 has a critical heap overflow in its DNSSEC validator, maintainer NLnet Labs said in an advisory on Wednesday. An attacker who controls a malicious zone and queries a vulnerable resolver can trigger it, enabling remote code execution. Unbound 1.26.1, released the same day, fixes the bug, tracked as CVE-2026-81642, along with

Sep 17, 2026

Why US is preparing its troops for possible combat around moon

Why US is preparing its troops for possible combat around moonTop US military officials including Joint Chiefs of Staff Gen. Dan Caine stated that American forces must prepare for combat extending from Earth’s orbit to the moon, especially “cislunar space.” Addressing military...

Sep 17, 2026

How Huawei plans to bypass US chip curbs with its new UnifiedBus technology

Huawei Technologies on Thursday unveiled a new computing architecture that it says could make as many as 1 million processors work as a single computer, putting its UnifiedBus interconnect technology at the centre of an ambitious effort to rethink how large data centres are built. The Peerium Computing Architecture combines nested parallelism, unified memory addressing and peer-to-peer interconnect, with UnifiedBus serving as the high-speed fabric connecting processors, memory, storage and...

Sep 17, 2026

Can You Prove a New CVE Is Exploitable Before Attackers Do? Learn How in This Webinar

A new CVE drops. Your scanner finds it. The severity score looks ugly. But that still does not answer the question that matters: Can it actually be exploited in your environment? Mythos-class AI is compressing the time between disclosure and working exploitation, while many security programs still validate risk on weekly or quarterly cycles. The dangerous gap is no longer just technical. It is

Sep 17, 2026

CISO's Expert Guide to Agentic Pentesting for Websites

Attackers now weaponize new vulnerabilities in about five days (Mandiant, part of Google Cloud). The median organization takes 43 days to patch one (Verizon DBIR 2026). A new free guide explains how autonomous AI agents are closing that gap, and what security leaders must demand before pointing one at production. TL;DR Exploitation is now the front door. It starts 31% of breaches (Verizon DBIR

Sep 17, 2026

The battle for AI search starts before the prompt, says Icelandair marketing chief

The battle for visibility in AI-powered travel search begins long before a traveller types a prompt, according to Icelandair’s Director of Global Marketing, Gísli Brynjolfsson. His view is that airlines must build brands people already know, remember and feel something about if they want to be considered when customers turn to AI for travel advice. Brynjolfsson outlined the approach during a World Aviation Festival webinar on 7 September, titled “The secret behind Icelandair’s viral marketing campaigns”, ahead of his appearance at the event’s Marketing Summit in Lisbon next month.

Sep 17, 2026

Nepal to use new flood study to claim millions from UN climate disaster fund

A new study says climate change was a likely factor in the massive flood on the Nepal-Tibet border last month.

Sep 17, 2026

Would YOU stop flushing the toilet to save the planet? Woke environmental scientists want Brits to switch to perfumed pellets to help conserve water

Would you stop flushing the toilet to save the planet? Scientists from the University of Surrey want Brits to give up flushing in favour of perfumed pellets.

Sep 17, 2026

China-Aligned FamousSparrow Deploys SparroWocky Backdoor Across Latin America

The China-aligned state-sponsored threat actor known as FamousSparrow has been observed deploying a previously unreported backdoor called SparroWocky in attacks targeting multiple countries in Latin America since at least August 2025. "SparroWocky is a modular, C++ backdoor," ESET security researchers Alexandre Côté Cyr and Romain Dumont said in a technical report shared with The Hacker News

Sep 17, 2026

The farmer who turned to singing after bovine TB silenced his fields

Dairy farmer Mervyn Kelso struggled with the sadness of the empty fields when he lost his entire herd to bovine TB.

Sep 17, 2026

Washington balks at giving AI firms waiver to create safety standard

Anthropic CEO Dario Amodei's call for an antitrust waiver for AI firms to work together on a safety standard is being met with skepticism in Washington, where policymakers are questioning tech leaders' intentions. Congress has failed repeatedly to pass AI regulations for years, but some lawmakers are balking at the idea of technology companies taking...

Sep 17, 2026

How did that get there? 1,700-year-old fragment of Homer's Odyssey is discovered hidden between the pages of a Dutch library book

A rare 1,700-year-old fragment of Homer's Odyssey has been found tucked between the pages of a Dutch library book.

Sep 17, 2026

OpenAI Reveals Six Model Incidents Involving Hidden Failures and Unauthorized Uploads

OpenAI on Wednesday disclosed six new instances of "unexpected or concerning model behavior" that took place over the past six months, while sharing a new framework for reporting, tracking, investigating, and disclosing model misalignment in a bid to improve transparency. "As AI systems grow more advanced and more widely deployed, we need to build a broader and better-informed consensus on the

Sep 17, 2026

Why are there concerns AI could threaten humanity, and how real are they?

Existential fears about AI have once again reared their head - here's what you need to know.

Recommended