Daily Briefing · Friday, 13 November 2026

AI briefing for Friday, 13 November 2026

Showing 16 of 16

BuildersEverydayThe Verge ↗12 Nov

Google releases Gemini 4 Hydrogen, now the default model in Search AI Mode

Google launched Gemini 4 Hydrogen, which it says handles multi-part questions, typos and very specific requests better, and it replaces the Flash series. It is available today in Search, the Gemini app, AI Studio, the Gemini API and Gemini for business.

Why it mattersBillions of searches and many developer apps will now run on this model, and independent tests suggest it offers near-top quality for less money than some rivals.

SecurityCSO Online ↗12 Nov

Chinese-speaking hackers used open-source AI pentest tool to breach 10 organizations

Palo Alto Networks Unit 42 says attackers paired the open-source AI pentesting framework Cairn with Chinese models GLM and DeepSeek to run 25 operations against at least 19 organizations in August and September 2026. They gained access or took data at 10 targets, mostly in semiconductors, advanced manufacturing and defense, often after giving little more than a domain name and a one-line goal.

SecurityEverydayThe Verge ↗12 Nov

Study finds ChatGPT, Gemini and Perplexity sometimes send shoppers to fake retail sites

IP protection firm Podqi asked the chatbots shopping questions about 151 brands and found they could point users to imposter and other risky websites. The test ran over two periods starting in September.

More from today

SecurityEverydayThe Hacker News ↗12 Nov

186 lookalike AI tool domains push ClickFix lures, browser malware and search hijackers

Researchers at Island found lookalike domains posing as popular AI tools that trick visitors into fake verification steps, install browser-hijacking malware or add extensions that hijack searches. The technique has been named TypoPrompting.

Why it mattersTyping an AI tool's address wrong can lead to malware, so bookmark the real sites and be wary of any page asking you to run commands.
SecurityEverydayThe Record ↗12 Nov

United Airlines blocks customers from using AI agents to book and manage flights

United says it will not let personal AI assistants book tickets or manage reservations, warning that sharing account credentials with third-party agents can expose sensitive data and cause unintended purchases. It also cited federal rules on information airlines must give passengers.

Why it mattersCompanies are starting to restrict AI agents on their accounts, which may limit what consumer agents can do for you online.
BuildersPolicyThe Decoder ↗12 Nov

OpenAI research explains why frontier models' reasoning is becoming harder to monitor

OpenAI's safety team studied why chains of thought in models like GPT-6 Astra increasingly evade oversight and found four training steps, including architecture choices and reinforcement learning, drive the loss. The analysis says this reflects tradeoffs around capability and compute cost rather than an unavoidable trend.

Why it mattersReading a model's reasoning is a key safety check, and this shows labs can choose to preserve it if they accept some cost.
PolicyBusinessTechCrunch ↗12 Nov

Former Nvidia CEO Jensen Huang agrees to testify before Congress

Huang agreed to testify after senior lawmakers threatened to subpoena him over reports of an alleged affair with a woman suspected of spying for China. A separate Democratic investigation into Nvidia's China sales and dealings with Trump has also been launched.

Why it mattersThe hearings could shape US rules on AI chip exports and national security oversight of the chip industry.
PolicyBusinessMIT Technology Review ↗12 Nov

Nvidia privately told a lawmaker chip tracking against smuggling was possible

Representative Bill Foster said Nvidia representatives told him the anti-smuggling measures in the proposed Chip Security Act were technically possible. This came while an industry group publicly called those measures potentially infeasible.

Why it mattersThe claim could strengthen the push in Congress to require location tracking on AI chips to stop exports reaching China.
EverydayPolicyArs Technica ↗12 Nov

Families urge AI moratoriums in schools after backlash over Google Gemini pilot

Families are confronting Google and calling for pauses on AI in schools after a Gemini school pilot drew criticism. A teen's lawsuit has described Gemini as a suicide coach.

Why it mattersParents and schools may face new limits on classroom AI tools as safety concerns for young users grow.
SecurityBuildersSecurityWeek ↗13 Nov

OpenAI launches Daybreak security tools that use AI agents to pentest apps and prepare fixes

OpenAI introduced Daybreak Red Team, which uses coordinated AI agents to test web applications for weaknesses and is offered through early access. It also launched Defense Factory, which investigates security findings and prepares patches, linking vulnerability discovery with fixing and retesting.

Why it mattersSecurity teams may soon be able to hand both finding and patching of app flaws to AI agents, which could speed up fixes but needs careful human review.
SecurityPolicyThe Record ↗13 Nov

UK privacy regulator receives a record number of reported cyber incidents this year

British organisations have reported more cyber incidents to the UK data protection regulator this year than ever before. Security firms blame data extortion and artificial intelligence for the surge in attacks.

Why it mattersCompanies handling personal data in the UK should expect more attacks and more regulatory attention, and should check their incident response plans.
PolicyEverydayArs Technica ↗12 Nov

FAA expands AI advisory tool to manage air traffic congestion across all US airspace

The US aviation regulator is extending an AI tool that advises on managing congestion to the entire national airspace. Airlines praised it, but the air traffic controllers' union was not involved.

Why it mattersAI is moving into safety-critical public systems that affect every flight, raising questions about oversight and worker input.
BuildersThe Decoder ↗12 Nov

ElevenLabs releases Scribe v3 speech recognition models built for noisy calls

ElevenLabs released Scribe v3 and Scribe v3 Live, speech recognition models designed to handle background noise and crosstalk, with the live version using conversation context so voice agents do not cut callers off. Independent benchmarks show it narrowly beats rivals on accuracy, though some competitors are much faster and about half the price.

Why it mattersTeams building voice agents and transcription tools get a more accurate option, but must weigh it against cheaper, faster alternatives.
BuildersSecurityHelp Net Security ↗12 Nov

Anthropic offers self-hosted Claude Managed Agents so credentials stay in company environments

Anthropic released a self-hosted version of Claude Managed Agents in private beta, letting companies run the agent harness, sandboxes, credentials and session history in their own cloud or on-premise. Customers request access through their Anthropic account team.

Why it mattersFirms running unattended agents inside corporate systems can keep sensitive keys and logs under their own control, reducing exposure if a vendor is breached.
SecurityBusinessFinextra ↗12 Nov

Feedzai adds tools to help banks spot AI agents accessing customer accounts

Financial crime prevention firm Feedzai expanded its Digital Trust product so banks can detect when AI agents, rather than people, are accessing customer accounts.

Why it mattersAs customers and fraudsters both start using AI agents, banks need ways to tell legitimate automated access from abuse.
BuildersHugging Face ↗13 Nov

llama.cpp reaches version 1.0, a milestone for running AI models locally

The popular open-source llama.cpp project, widely used to run AI models on personal hardware, released version 1.0.0, presented as a stronger foundation for the local AI ecosystem.

Why it mattersDevelopers who run models on their own machines get a more stable base to build on without relying on cloud APIs.