Jan 16, 2026

AI News Roundup: OpenAI Launches $8 ChatGPT Tier with Ads, Musk Trial Set for April

OpenAI just revealed its plan to turn ChatGPT into an advertising platform. The company launched ChatGPT Go, an $8/month subscription tier, while simultaneously announcing it will begin testing ads in the coming weeks. Free and Go users will see targeted ads; Plus, Pro, Business, and Enterprise users will remain ad-free.

The move signals a dramatic shift in OpenAI’s business model - and comes just as a federal judge ruled that Elon Musk’s fraud lawsuit against OpenAI will proceed to trial, with potential damages ranging from $1 billion to over $50 billion.

Here’s everything that matters from January 15-16, 2026.

The Big Story: OpenAI Embraces Advertising Model

OpenAI made two major announcements that fundamentally reshape how hundreds of millions of people will interact with ChatGPT. First, the company launched ChatGPT Go globally at $8/month - positioned between the free tier and the $20/month Plus subscription. Go subscribers get 10x more messages than free users, file uploads, image creation, and unlimited access to GPT-5.2 Instant.

But the bigger news is what’s coming: targeted ads. OpenAI confirmed it will begin testing advertisements in the coming weeks for free and Go tier users. The ad-free experience becomes a premium feature reserved for Plus subscribers ($20/month) and above.

The implications are significant. ChatGPT has become a primary information source for hundreds of millions of users. Introducing advertising creates incentives that could influence how the AI presents information, which products it recommends, and how it frames answers. OpenAI insists ads will be clearly labeled and won’t compromise response quality - but the same promises have been made by search engines before.

This also represents a major new revenue stream. With 230 million weekly users asking health questions alone, the advertising opportunity is enormous. OpenAI simultaneously launched ChatGPT Health and OpenAI for Healthcare - HIPAA-compliant products now rolling out to Boston Children’s Hospital, Cedars-Sinai, and Stanford Medicine Children’s Health.

Sources: OpenAI Official, TechCrunch, CNN Business

Today’s Top Stories

Musk v. OpenAI Fraud Trial Set for April 2026

U.S. District Judge Yvonne Gonzalez Rogers rejected OpenAI and Microsoft’s requests to dismiss Elon Musk’s fraud claims in a 28-page ruling. The judge found “ample evidence” supporting Musk’s allegations that OpenAI violated its nonprofit commitments.

Trial is scheduled to begin April 27, 2026, with potential damages ranging from $1 billion to over $50 billion - representing the value Musk claims he lost by not participating in OpenAI’s commercialization. OpenAI warned investors to expect “deliberately outlandish claims” from Musk during the proceedings.

This will be the highest-stakes AI trial in history, potentially determining whether OpenAI’s for-profit transformation was legally proper.

Sources: Bloomberg, TechCrunch

Anthropic Approaches $350 Billion Valuation

Anthropic is preparing to close a $10 billion funding round at a $350 billion valuation, led by Coatue Management and Singapore’s sovereign wealth fund GIC. This comes just three months after the company’s $13 billion Series F at $183 billion - nearly doubling its valuation in a quarter.

The company also announced major leadership changes: Mike Krieger is stepping away as Chief Product Officer to co-lead Anthropic’s Labs team, while Ami Vora takes over as head of product alongside newly appointed CTO Rahul Patil. Additionally, Claude Cowork - which allows Claude to access your file system and navigate websites - expanded from Max subscribers ($100+/month) to all Pro subscribers ($20/month).

Sources: ContentGrip, Anthropic Official

DeepMind CEO: China “Months” Behind US in AI

Google DeepMind CEO Demis Hassabis told CNBC that Chinese AI models might be “a matter of months” behind U.S. and Western capabilities - a much narrower gap than previously assumed. However, he questioned whether China can “innovate beyond the frontier,” suggesting they’ve demonstrated an ability to catch up but not yet to create fundamentally new breakthroughs.

Hassabis also revealed he talks to Google CEO Sundar Pichai “every day”, calling DeepMind “the engine room” of Google’s AI efforts. He said the team “got into our groove” with Gemini 2.5 in March 2025, followed by the highly praised Gemini 3 launch in November.

Source: CNBC

EU and UK Launch Major Probes into Grok Deepfakes

The deepfake crisis engulfing xAI has escalated dramatically. The EU, UK (Ofcom), California, and Canada launched investigations into Grok’s deepfake capabilities, while Malaysia and Indonesia have already imposed temporary bans.

Ashley St. Clair, the mother of one of Elon Musk’s children, sued xAI alleging Grok generated sexually explicit deepfake images of her without consent. xAI confirmed Grok will no longer edit “images of real people in revealing clothing” following the backlash.

Sources: CNN Business, Financial Content

Quick Hits

OpenAI Infrastructure: OpenAI announced a $10B+ multi-year deal with Cerebras for 750 megawatts of ultra-low latency compute - the world’s largest high-speed AI inference deployment. The company is also lining up U.S.-based suppliers for data centers, robotics, and consumer devices.
Database AI Boom: ClickHouse raised $400M at $15B valuation, more than doubling its value as AI agent applications drive database demand.
Funding Frenzy: Parloa raised $350M at $3B valuation; Replit nears $400M raise at $9B valuation; Higgsfield AI raised $80M at $1.3B valuation.
Meta Goes Nuclear: Meta announced landmark agreements with Vistra, TerraPower, and Oklo to unlock up to 6.6 GW of nuclear power for AI infrastructure - making it one of the largest corporate nuclear energy purchasers in U.S. history.
TSMC Record Quarter: TSMC delivered another estimate-beating quarter with profit up 35% YoY. Capital expenditure expected to reach $52-56B in 2026.
MIT’s 2026 Breakthroughs: MIT Technology Review’s annual list features mechanistic interpretability, AI companions, and generative coding - AI now writes 30% of Microsoft’s code and over a quarter of Google’s.
Security Alert: Security researchers disclosed “Reprompt” attack allowing data exfiltration from Microsoft Copilot with a single click. Separately, Urban VPN Proxy was found harvesting conversations from 10 AI platforms by default.

What This Means

Today marks a turning point in how we interact with AI. OpenAI introducing ads to ChatGPT transforms the world’s most popular AI assistant from a subscription product into an attention-monetization platform - the same model that shaped social media’s trajectory. The simultaneous Musk trial ruling ensures that questions about OpenAI’s nonprofit-to-profit transformation will be litigated in public view with billions at stake. Meanwhile, Anthropic’s rapid valuation growth to $350B and DeepMind’s admission that China trails by only “months” show this is a race where advantages are measured in weeks, not years. The regulatory crackdown on Grok’s deepfakes across multiple jurisdictions signals that AI companies can no longer treat safety as an afterthought - the political will for enforcement has finally materialized.