2026
InsightFinder raised $15M for diagnosing failures in AI-agent systems
Startup InsightFinder received $15M for developing diagnostic tools not only for AI models, but for the entire technical stack in which agen
Anthropic Releases Claude Opus 4.7 — New Flagship for Complex Coding
Anthropic introduced Claude Opus 4.7 — the company's most powerful publicly available model, featuring improved coding, image analysis, and
Gemini now creates personalized images from your Google Photos
Google expanded Personal Intelligence: Gemini uses your photos from Google Photos and the Nano Banana 2 model to generate images tailored to
OpenAI Updated Codex: Desktop Management, Image Generation, and Task Memory
OpenAI has expanded Codex's capabilities: the agent can now manage applications on the macOS desktop, generate images, and retain context fr
Google adds split-screen to AI Mode: websites now appear alongside chat
Google updated AI Mode in Chrome Desktop — now when you click a link, the website opens alongside the AI chat instead of in a separate tab,
Google updated AI Mode: sources now open next to the chat
Google added a source-viewing mode to AI Mode without leaving the chat — the website opens on the side, and you can immediately ask question
Google updated AI Mode in Chrome — now it doesn't disappear when switching tabs
Google changed the behavior of AI Mode in Chrome: now the chat search remains active throughout the entire session, not disappearing when sw
UK Launches $675 Million Sovereign AI Fund to Support Domestic Startups
The UK government is creating a $675 million sovereign fund to finance domestic AI companies and reduce the country's technological dependen
Musk v. Altman Court Case: Jury to Assess OpenAI's Fidelity to Its Mission
A jury will soon deliver a verdict in the Musk v. Altman case and determine whether OpenAI has strayed from its founding goal — to create AG
OpenAI Updated Codex: Agent Now Takes Control of Desktop
OpenAI has rolled out a major update to its Codex programming agent tool, adding new desktop management capabilities.
Physical Intelligence Introduces π0.7 — a Robot Brain Capable of Mastering Unfamiliar Tasks
Startup Physical Intelligence released the π0.7 model, which handles tasks it was never trained on — a step toward universal general-purpose
Upscale AI in talks for third round at $2 billion valuation seven months after launch
AI infrastructure startup Upscale AI is discussing a new funding round at a $2 billion valuation — already its third in the company's first
Luma AI launches Wonder Project studio: Moses film with Ben Kingsley to premiere on Prime Video
Luma AI is opening Wonder Project, an AI production studio with a religious focus. Its first project, a Moses film with Oscar winner Ben Kin
Factory raises $150M from Khosla Ventures, reaches $1.5B valuation
Three-year-old startup Factory has closed a $150M round led by Khosla Ventures, valuing the company at $1.5B — it builds AI agents for enter
The “AI Is Inevitable” Trap: Have We Reached Peak Hype Around Artificial Intelligence
Footwear maker Allbirds declared itself an AI company and saw its value rise sevenfold in a day — an episode that became a symbol of the cur
Dairy Queen rolls out voice AI at dozens of drive-thrus across the US and Canada
Ice cream chain Dairy Queen is launching Presto's AI chatbot at drive-thrus to speed up service and increase average order value.
Tokenmaxxing and the chasm: OpenAI buys up everything while Anthropic hides a model
Tokenmaxxing, OpenAI acquisitions, a footwear brand as AI infrastructure, and an Anthropic model “too powerful for the public” — the gap bet
Anthropic launches Claude Design — quick visuals for non-designers
Anthropic has released Claude Design, a standalone tool for quickly creating visuals aimed at founders and product managers without design s
OpenAI buys up companies, Anthropic withholds a model — the AI divide between insiders and everyone else is widening
While OpenAI buys up fintech apps and talk shows, a shoe manufacturer rebranded itself as AI infrastructure, and Anthropic admitted: the new
Wired: Editorial offices secretly deploy AI assistants for article writing — and remain silent about the consequences
Newsrooms increasingly entrust AI with drafting articles under the guise of work acceleration, but the real cost of this efficiency turns ou
Anthropic reconciles with Trump: new Claude Mythos model for cybersecurity
After two months of public conflict, Anthropic may restore relations with the Trump administration thanks to the specialized Claude Mythos P
Sora Team Lead Bill Pebbles Leaves OpenAI After Project Shutdown
After OpenAI shut down its Sora video generator, team lead Bill Pebbles announced his departure from the company.
Kevin Wale and Bill Pebbles Depart OpenAI: Company Shuts Down Sora and Research Division
OpenAI loses its product director and Sora project lead — the company sharply pivots toward enterprise AI solutions and abandons consumer ex
Schematik — "Cursor for Hardware": Anthropic Considers Investment in Hardware Startup
Schematik enables designing physical devices through dialogue with AI — like Cursor for code, but for electronics. Anthropic wants to invest
App Store Experiences New Launch Boom — Appfigures Links Growth to AI Development Tools
According to Appfigures, the number of new apps in the App Store surged sharply in 2026 — analysts attribute the main cause to the prolifera
Cerebras Startup Files IPO Application: AWS and OpenAI Deals Worth $10 Billion
AI chip manufacturer Cerebras Systems has filed documents for IPO. The company has reached agreements with Amazon Web Services and OpenAI, w
Tesla Expands Robotaxi to Dallas and Houston — Service Extends Beyond Austin
Tesla launched driverless robotaxi in two new cities — Dallas and Houston — publishing video of electric vehicles with no drivers or operato
Uber shifts to asset ownership strategy — what changes for the taxi market
Uber abandons the "platform only" model and begins building its own asset fleet—vehicles, infrastructure, robotaxis.
12-month window: AI startups live while OpenAI hasn't reached their niche
Most AI startups exist in niches that major foundational models haven't reached yet—but this is temporary, and founders understand this perf
OpenAI and two existential challenges: how acquisitions should save the company
Analysts from the Equity podcast analyzed OpenAI's recent acquisitions and assessed whether they can solve two key crises threatening the co
Zuckerberg and Dorsey want to clone themselves with AI — to lead from anywhere
Two Silicon Valley executives are building AI management systems that would let them be everywhere at once — and deepen corporate control.
NSA uses Anthropic's classified Mythos model despite conflict with the Pentagon
The US National Security Agency is secretly working with Anthropic's closed AI model, ignoring the company's public conflict with the Pentag
Epic Games makes AI character creation tool available to Fortnite developers
Epic Games launched Conversations, a tool that lets map creators build NPCs with unscripted AI-based dialogue.
Google launches Gemini in Chrome browser in seven new Asia-Pacific countries
Google expands access to Gemini in Chrome to Australia, Indonesia, Japan, Philippines, Singapore, South Korea and Vietnam.
Anthropic Raises $5B from Amazon in Exchange for $100B in Cloud Spending
Amazon invested another $5 billion in Anthropic, and the AI company committed to spending $100 billion on AWS—the largest circular deal in A
Honor robot completes half-marathon in 50:26 — 7 minutes faster than world record
Autonomous humanoid robot Honor completed 21 km in 50 minutes 26 seconds, beating the world human record by 7 minutes — without remote contr
Yelp Transforms AI Assistant Into a Digital Concierge With Booking
Yelp Assistant now sits at the center of the app: recommends restaurants, answers questions and books tables — all in one conversation witho
Medical Student Made Thousands Selling Fake AI-Generated Conservative Girl
A medical school student created an AI image of a young MAGA-style conservative and sells her photos and videos without warning buyers that
John Ternus will replace Tim Cook at Apple — and immediately confront its AI failure
Apple announced: John Ternus, a 25-year hardware veteran, will become CEO on September 1. His main challenge is AI, where Apple is still hop
Startup GRAI believes AI should help fans remix music, not replace artists
AI startup GRAI offers a different view of AI's role in music: instead of generating tracks from scratch, it builds tools for collaborative
YouTube expands AI likeness detection to celebrities — now they can fight deepfakes
YouTube is giving celebrities and their representatives access to an AI tool for automatically finding deepfakes — performers will be able t
Bond — a new social network that uses AI to wean you off doomscrolling
Startup Bond has created a platform where AI actively encourages users to close the app and return to the real world — instead of keeping th
Latitude unveils Voyage, a platform for building AI-RPGs with living NPCs
Latitude, the creator of AI Dungeon, has announced Voyage, a platform for creating full-fledged RPG worlds with characters powered by genera
At the FTC's request, Clarifai destroyed 3 million OkCupid user photos leaked for AI training
In 2014, dating platform OkCupid secretly handed millions of user photos to Clarifai for facial recognition — the images have now been destr
Sam Altman criticized Anthropic's new model Mythos: "fear-based marketing"
The head of OpenAI called the positioning of Anthropic Mythos, a model tailored for cybersecurity, fear-based marketing rather than a demons
YouTube expands deepfake protections: Hollywood stars will be able to remove AI clones
YouTube has launched a deepfake detection tool for celebrities — they will be able to track AI videos featuring them and request their remov
OpenAI explained why Codex Security does not use traditional SAST analysis
OpenAI explained why Codex Security moved away from classic SAST scanners in favor of AI reasoning over code constraints — with minimal fals
OpenAI released GPT-5.4 mini and nano — compact models for agentic tasks and coding
OpenAI announced GPT-5.4 mini and nano — lighter versions of its flagship model, optimized for agentic systems, coding, and high-throughput
OpenAI Japan unveils Japan Teen Safety Blueprint to protect minors
OpenAI Japan announced the Japan Teen Safety Blueprint — a package of measures to protect teens, including enhanced age verification, parent
OpenAI explains how it tracks signs of misalignment in AI coding agents
OpenAI outlined a chain-of-thought monitoring system the company uses to detect signs of misalignment in coding agents used in real-world se
OpenAI revealed Sora 2’s safety principles: from deepfakes to content labeling
OpenAI described a layered safety system built into Sora 2 and the new social platform — from model filters to technical video attribution u
ChatGPT launches visual shopping powered by Agentic Commerce Protocol
OpenAI has added a full shopping engine to ChatGPT: visual product cards, option comparison, and direct integration with stores through a ne
OpenAI released teen safety prompt policies for developers of gpt-oss-safeguard
OpenAI has opened a set of ready-made teen safety policies based on gpt-oss-safeguard to developers — age-based moderation without rewriting
OpenAI launches Safety Bug Bounty for vulnerabilities in agentic AI systems
OpenAI has opened a rewards program for researchers who find vulnerabilities specific to AI: prompt injections, data leaks and attacks on ag
OpenAI explained how Model Spec is structured — the code of conduct for its models
OpenAI openly described the principles that govern its models' behavior — from the hierarchy of priorities to the delineation of operators'
OpenAI raises $122 billion to expand frontier AI and compute capacity
OpenAI has closed a record $122 billion round — the funds will go toward next-generation compute, global expansion, and growth of its enterp
Gradient Labs provided every bank customer with a personal GPT-4.1 AI manager
Startup Gradient Labs uses GPT-4.1 and compact mini/nano models for AI agents that automate banking customer support in real time.
OpenAI launches fellowship program for independent AI safety researchers
OpenAI opens a pilot grant program for independent researchers working on AI safety and alignment to help develop the next generation of spe
OpenAI published a plan to protect children from AI risks
OpenAI introduced Child Safety Blueprint, a document with principles for the safe design of AI products to protect minors online.
OpenAI announced a new phase of enterprise AI: ChatGPT Enterprise and company-scale agents
OpenAI is moving from AI assistants to autonomous company-scale agents: at the center of this new phase are ChatGPT Enterprise, Frontier, Co