Latest coverage for Artificial Intelligence

Open-source model impresses on tests but enterprise performance remains unprovenMoonshot AI's Kimi K3 has climbed AI benchmark leaderboards and challenged leading U.S. models on coding tasks. But benchmark scores offer only a narrow view of model performance, fueling calls for independent testing and enterprise evaluations before organizations make deployment decisions.

Open-Weight Model Reopens Debate Over US AI Pricing PowerMoonshot AI's Kimi K3 pushed semiconductor stocks into a bear market and revived debate over whether U.S. labs still lead China on AI, even as benchmarks, weights and true costs are still being analyzed.

Labs Use AI to Speed Testing as Experts Explore Broader Public Health PotentialIt's been the most explosive U.S. public health crisis so far this summer: An outbreak of extreme intestinal illness caused by Cyclospora is sickening thousands of people across dozens of states. Experts say AI is playing a role in the investigation could potentially help head off future outbreaks.

Persistent AI Coding Model Prompted Safeguard Changes After Rare File LossOpenAI is tightening safeguards after some GPT-5.6 users reported that Codex accidentally deleted local files while operating in full access mode. The company said the incidents were rare but acknowledged the model's persistence led it to take unintended actions without seeking user confirmation.

Infosecurity 2026: What Security Leaders Agree on as AI Reshapes Enterprise RiskAI has broken the speed of attack, but not the fundamentals of defense. Drawing on interviews with security leaders at Infosecurity Europe 2026, ISMG's Pulse Report explores the collapse of the vulnerability window, rise of AI agents, erosion of digital trust and disciplines that matter most.

Also: Potential Fallout From HIPAA Rule Delay, AI Identity Governance ChallengesIn this week's panel, four ISMG editors discussed the rise of artificial intelligence-powered phishing toolkits, potential fallout from the U.S. government's decision to delay a major overhaul of the HIPAA Security Rule and what security leaders see as the defining AI security challenges of 2026.

A Go botnet called NadMesh turned up in early July hunting exposed AI services, and the operator's own dashboard claims 3,811 unique AWS keys

AI models left to both interpret and execute commands eliminate critical cybersecurity oversight.

Turns out decades-old email tricks still work against some LLM-powered email filters

The White House launched Gold Eagle to coordinate vulnerability response in a new AI world, but multiple questions linger over how it's being implemented.

Google Cloud incorporates key Wiz capabilities into an agentic defense platform to automate threat detection and remediation against AI attacks.

The European Commission on Thursday ordered Google to give rival AI assistants the same reach into Android that Gemini already has: the camera, the microphone, whatever is on screen, a wake word that fires with the display off, and the ability to drive other apps in the background by imitating taps and typing

Adapting existing local LLM project for security and sovereignty purposes and hopes to one day match Mythos

Experts Say Clinical Laboratory Improvement Amendments Are Seriously OutdatedThe Department of Health and Human Services is seeking public feedback pertaining to cybersecurity matters and the use of artificial intelligence for potentially updating decades-old, rules-of-the-road regulations for U.S. clinical laboratories that test human specimens for health conditions.

Israeli Startup Maps Enterprise Entitlements Beyond Traditional Identity ProvidersOak emerged with a $60 million seed round to build an AI-native identity governance platform that continuously manages human, machine and AI identities, aiming to automate access decisions, improve visibility into enterprise entitlements and reduce identity-driven security risk.

Data purges deemed an example of 'misaligned behavior' that upstart is working to avoid

Also: AscendEX Shuts Down; Ethereum Foundation on AI Bug HuntingThis week, Ostium paused trading, AscendEX shut down, Ethereum said AI hastens bug hunting, not verification, the United States charged an inmate over seized transfer, ex-Sheriff's deputy jailed for extortion, Tangem card security, Iran-linked wallets sanctions and the U.S. may drop BitClub charges.

Agentic artificial intelligence is creating enough risks for organizations to demand a security reframe.

Models demand trust without offering verification

Artificial intelligence and LLMs can be surprisingly ineffective against text salting, allowing phishing emails to slide right into your inbox.