TechCrunch
AI news source. Articles are auto-selected and adapted by Hamidun News editors.
Latest publications
AI in Advertising: Google Requires Disclosure of AI-Generated Content
Google is expanding its requirement to disclose AI content beyond political ads — now all advertisers must report the use of synthetic content.
AI Agent for Business: How Lyzr Raised $100M Using Its Own Agent
Startup Lyzr, a developer of AI agents for enterprises, raised a $100M round, letting its own agent run the fundraising process as a proof of concept.
SambaNova raises $1B at $11B valuation, rejecting Intel offer
AI chipmaker SambaNova closed a $1B Series F round at an $11B valuation, rejecting Intel's lower $1.6B acquisition offer.
Fastest-Growing AI Startups Accelerate Revenue Growth Momentum
Among fast-growing AI startups, standout companies are scaling revenue significantly faster than competitors and industry averages.
Game data, not LLM: how startup General Intuition sees the path to AGI
General Intuition, a startup backed by Jeff Bezos, argues that achieving AGI requires game data, not just language models like ChatGPT and Claude, which fail to understand the physics of space and time.
SpaceXAI releases Grok 4.5: Elon Musk calls the model Claude Opus level
Elon Musk's SpaceXAI has unveiled Grok 4.5, promising a cheaper and more efficient alternative to competitors' powerful models.
OpenAI adds simultaneous listening and speaking to voice mode for translation
OpenAI unveiled an updated voice mode: the AI can now listen and speak at the same time, enabling real-time speech translation without pauses.
OpenAI Launches GPT-5.6 and New Model Family with Enhanced Security
On July 9, 2026, OpenAI unveiled a new family of language models centered on GPT-5.6. The update promises improvements in cybersecurity and other AI development areas.
Character.AI Launches Interactive Video Series Where Viewers Chat with Characters
Character.AI added interactive micro-drama features — short video series where viewers chat with characters through an embedded chat, ask questions, and replay scenes.
Ollama Raises $65 Million and Nearly Reaches 9 Million Users for AI Tool
Open-source tool Ollama for running AI locally has raised $65 million and nearly reached 9 million users, cementing its popularity among developers.
Nvidia Loses Ground: Memory Shortage Reshapes the AI Computing Market
Nvidia shares fell 15% from their May peak while Micron nearly tripled in value: GPU scarcity has eased, making memory the new bottleneck for AI data centers.
Meta to begin production of its own AI chips in September 2026
Meta announced the launch of production of its own AI chips in September 2026, using a modular architecture for rapid adaptation to the development of artificial intelligence. *Meta is recognized as an extremist organiz
How authorities decided that OpenAI’s advanced model was safe to release
A July 9, 2026, TechCrunch report examines how authorities assessed the safety of OpenAI’s advanced model, but the details of the dialogue with the companies remain unknown.
Paris-based voice AI startup Gradium raises $100 million with Nvidia backing
Paris-based startup Gradium, which develops AI-powered speech synthesis and processing technologies, has received $100 million in seed funding and will open an office in the San Francisco Bay Area.
Amazon closes Mechanical Turk to new clients: end of the crowdsourcing AI era
Amazon has stopped accepting new clients to Mechanical Turk, the platform that for two decades provided the AI industry with labeled data for model training.
Station F opens new intake for AI-accelerator F/ai for European startups
Parisian startup campus Station F, founded by Xavier Niel, announced a new intake for AI-accelerator F/ai — a specialized program for the most promising European AI companies.
Why Open AI's Growth Doesn't (Yet) Threaten Anthropic: A Lifecycle Analysis
TechCrunch explains the paradox: open models set records, but Anthropic continues to grow — because each player occupies its own phase of a single lifecycle.
Meta launched Muse — an AI image generator for ads, interior design, and content creators
Meta officially released Muse — a generative image model focused on advertising campaigns, interior design visualization, and content creator tools within Facebook and Instagram. *Meta has been recognized as an extremist
French startup ZML released free tool to accelerate AI inference on any chips
ZML, endorsed by Turing Award laureate Yann LeCun, has opened free access to ZML/LLMD — software for accelerating AI model inference on different types of chips.
Former OpenAI Product Director Kevin Weil Joins Stoke Space Board
Kevin Weil, one of OpenAI's key executives, has joined the board of Stoke Space — a startup developing fully reusable rocket launchers.
Microsoft cuts 4,800 employees: Xbox and commercial sales hit hardest
On July 7, 2026, Microsoft laid off 4,800 people — 2.1% of its global workforce. Xbox and commercial sales teams were hit hardest amid concerns about AI-driven optimization.
Google started training AI on your data — here's how to turn it off
Google changed its privacy settings: your data is now used by default to train AI models. Here's how to opt out manually.
Reddit uses language models against AI-generated spam — largely created by those same LLMs
Reddit has started using large language models to fight AI-generated spam — the platform is forced to battle AI with the same tools used to attack it.
2026 chronicle: which tech companies officially cited AI as the reason for layoffs
TechCrunch maintains an updated list of major tech companies that in 2026 are officially citing AI as a factor in mass layoffs.
Vercel CEO Guillermo Rauch: 1 trillion tokens a day and the split between AI models and agents
Vercel handles 6 million deploys a day — half are launched by agents, and 1 trillion tokens pass through the AI Gateway. CEO Guillermo Rauch explained why production AI requires a multi-model architecture.
AI technically executed the first ransomware attack, but a human selected the target and provided access
An AI agent technically executed a real ransomware attack for the first time, but a human operator chose the victim, set up the infrastructure, and supplied the credentials — it was not as autonomous as the headlines sug
Discord admits AI moderation error: hundreds of accounts blocked over harmless photos
Discord confirmed that a bug in AI moderation had been blocking accounts over ordinary images since May 2026 — another 200 users were affected over the July 5–6 weekend alone.
Microsoft bets on in-house AI models to cut costs
Microsoft has joined the trend of cutting AI spending and plans to significantly expand the use of its own models, reducing its reliance on third-party providers.
Microsoft: over 20 million paying Copilot users — and they actually use it
On April 29, 2026, Microsoft revealed: Copilot has over 20 million paying users, and engagement continues to grow — despite long-standing skepticism that no one actually uses the company's AI assistant.
Nadella: Microsoft will offer OpenAI technologies to Azure customers without additional costs
Microsoft has the right to resell OpenAI technologies to Azure enterprise customers without payments to the developer — and Satya Nadella intends to "use this to the fullest."