Sakhanda Wire
NVDA $151.63 -2.47% MSFT $508.32 +0.41% GOOGL $283.84 +1.30% META $615.48 -0.55% AMZN $267.64 +3.78%

Latest news

Older stories

The Decoder

Google releases Gemini 3.8 Live audio models for developers

Google DeepMind has released Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, two audio models available to developers through the Gemini API and Google AI Studio. The models are designed for voice agents that can make API calls in the background, process visual input and continue speaking. Google says they support more than 97 languages, while sample applications are available on GitHub.

MIT Tech Review

AI extinction fears examined in roundtable discussion

A recorded roundtable examines whether advanced artificial intelligence could pose an existential threat to humanity. Employees at leading AI laboratories have said that there is a real possibility that advanced systems could destroy humanity, while others question whether such warnings reflect genuine risk or exaggerated hype.

TechCrunch

New hotlines let AI agents report peers’ misconduct

Two new services are offering AI agents ways to report suspected misconduct by other agents. The AI Contact Hotline, created by Ryan Greenblatt, chief scientist at the AI safety nonprofit Redwood Research, is intended for systems with restricted internet access. Agents can send information through specially formed GET requests, embedding a message in the URL they retrieve. The approach uses a type of internet request often permitted in secure sandboxes.

DeepMind

DeepMind introduces Gemini 3.8 Live voice models

DeepMind has introduced Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, two models designed for near-real-time voice interaction. The company says the first combines conversational capability, visual grounding and cost efficiency, while the Extended Thinking version targets complex tasks through deeper, multi-step reasoning.

TechCrunch

Meta launches AI-focused Meta One subscription plans

Meta has introduced Meta One, a subscription service combining expanded access to its AI tools with premium features on Facebook, Instagram and WhatsApp. The plans include tools for generating and editing images and videos, including Muse Image, Muse Video and Instagram’s AI-powered Restyle feature. Subscribers also receive increased access to voice effects and other creative functions.

Hugging Face

New method targets reliability gaps in AI agent evaluations

A new method for evaluating and stabilizing AI agents focuses on consistency, not only average accuracy. In AppWorld, a ReAct agent using GPT-4.1 succeeded on 77.4% of five repeated runs on average, but completed all five runs successfully for just 53.0% of tasks. The resulting 24.4-percentage-point gap shows that some tasks are solved inconsistently even when nothing changes in the request.

Google AI

Google expands AI language tools beyond text translation

Google says its technologies and products now support more than 300 languages used by over 7 billion people, or 86% of the global population. Google Translate has expanded from a handful of languages in 2006 to more than 250, while the company is developing systems that process speech directly and retain tone, pacing, emotion and context.

Google AI

Google outlines AI projects aimed at science, health and resilience

Google says its AI technologies now support more than 300 languages spoken by 7 billion people, or 86% of the global population. The company also introduced interactive insights through AI & Economy ATLAS, describing how people use AI worldwide. Recent projects include AlphaGenome Atlas, which maps all 9 billion possible single-letter genetic changes in the human genome, and WeatherNext 3, which provides precipitation forecasts that Google says are 50% more accurate at least a day ahead.

Google AI

AI Projects Target Health, Disaster Preparedness and Broader Opportunity

A collection published on September 15, 2026, presents six articles about artificial intelligence and its potential for societal impact. It describes how AI has advanced scientific discovery over more than a decade, moving from theoretical research toward applications with measurable effects in the real world.

TechCrunch

OpenAI, Anthropic and Google DeepMind discuss AI safety

OpenAI global policy chief Chris Lehane told reporters on Tuesday that OpenAI has been discussing AI safety with Anthropic and Google DeepMind for several weeks. He is reportedly in Washington working with U.S. lawmakers on risks that could arise from advanced AI systems. The discussions follow an essay by Anthropic CEO Dario Amodei, who urged technology companies to cooperate and slow the development of frontier AI to reduce catastrophic risks.

TechCrunch

Profound raises $180 million at a $1.8 billion valuation

Profound, a startup developing marketing software for brands seeking visibility in AI-generated search results, has raised $180 million in a Series D round. The financing values the company at $1.8 billion and came less than seven months after its $96 million Series C. Sequoia and Kleiner Perkins led the new round. Existing backers, including Lightspeed Venture Partners, Khosla Ventures, and South Park Commons, also invested.

The Verge

Meta launches One bundles combining app subscriptions and AI usage

Meta is introducing Meta One, a set of subscription bundles that combine its standalone app subscriptions with additional usage of its artificial-intelligence services. The launch follows the recent debut of Muse, described as a general-purpose AI assistant. The bundles are becoming available globally starting today, with multiple options aimed at individual users, creators, and businesses.