Marvell’s 115 Billion Lumentum Deal Leads AI News

Marvell’s planned 115 billion all-stock acquisition of Lumentum is the day’s biggest corporate development, combining two companies positioned around optical networking for AI data centers. Anthropic also updated its IPO filing with sharply higher quarterly revenue and much larger compute commitments, while OpenAI announced new Abu Dhabi agreements and hired a former Army secretary for a national-security role. Elsewhere, lawsuits over AI scraping, limits on Grok Bot usage, and a new restriction on Claude Code’s game-cloning capabilities added to the day’s consequential follow-ups.

Marvell Bets 115 Billion on Lumentum’s Optical Networking

Marvell said it would acquire Lumentum in an all-stock transaction valued at 115 billion, bringing together two major AI-hardware companies. The deal is centered on photonics and optical networking, technologies increasingly used in AI data centers. The New York Times reports that the companies described the combination as a major bet on that market.

Investors initially reacted cautiously: Marvell shares fell as much as 7% and Lumentum shares fell 12%, according to Business Insider. The response reflects questions about whether the expected revenue and cost synergies will justify the price as scrutiny of AI-data-center economics increases.

Anthropic’s IPO Filing Shows Revenue Growth and a Huge Compute Bill

Anthropic’s amended S-1 reported nearly 17 billion in third-quarter revenue, more than 5 billion above the prior quarter and roughly 12 times the year-earlier figure, according to Business Insider’s account of the filing.

The filing also showed a 567 million operating loss for the quarter. Anthropic said the loss was primarily driven by 5.0 billion of compute under short-term commitments, which it described as more expensive than comparable multiyear arrangements. The company disclosed about 895 billion in committed compute and infrastructure spending, with nearly half expected by the end of 2030. Those figures give investors a more detailed view of the spending obligations behind the company’s rapid growth.

OpenAI Expands Its Abu Dhabi Agreements

OpenAI signed four agreements in Abu Dhabi spanning energy, government, and investment, according to CNBC. The agreements were announced as Sam Altman met the emirate’s crown prince.

Bloomberg reported that the arrangements involve deploying OpenAI technology across energy companies, investment firms, and government agencies. The deals deepen the company’s relationship with the UAE’s state-backed investors as OpenAI seeks funding from the country.

OpenAI Creates a National-Security Modernization Role

OpenAI hired former U.S. Army Secretary Dan Driscoll for a newly created position titled “Senior Counselor for National Security Modernization,” according to CNBC.

The company said Driscoll will help expand the use of artificial intelligence for national and cyber security, Bloomberg reported. The appointment adds a senior government figure to OpenAI’s efforts in national-security applications.

The New York Times Sues Brave Over Alleged AI Scraping

The New York Times and other publishers sued Brave Software, alleging that the privacy-focused browser company scraped their content and distributed it to customers, including Anthropic and Perplexity. TechCrunch describes the complaint as targeting Brave’s alleged use of copyrighted material and unlawful distribution.

According to The Verge’s report, the publishers also allege that Brave monetized a scraped index by providing AI companies with paid access to real-time online content. The claims remain allegations in litigation, but the case adds another dispute over how web indexes can supply material to AI systems and users.

Grok Bot Usage Limits Reflect Serving-Cost Pressure

Elon Musk said the cost of serving SpaceX’s Grok Bot was “unsustainably high relative to revenue” after users complained that they had reached their usage limits. Business Insider reported that SpaceX introduced or tightened weekly allowances across Grok Bot’s paid tiers.

The service launched in August with plans ranging from free access to 200 a month and usage limits that refresh weekly. SpaceX said last month that Grok Bot had more than 800,000 paid weekly users, but the company’s comments show that rapid usage growth has brought direct pressure on the cost of operating the agent.

Anthropic Restricts Claude Code Game Cloning

Anthropic updated Claude Code’s instructions to tell the tool not to recreate specific existing video games, even if a user renames the game or calls it original. Ars Technica reported that the rule appeared in a Claude Code release and that Opus 5.5 was also reported to block some reverse-engineering tasks.

The company’s restriction still allows users to build original games, games in a genre, and projects based on their own designs and assets. Anthropic confirmed the limits and said it was iterating to find the right balance, according to Tom’s Hardware.

The Open Questions Behind Today’s Announcements

Marvell and Lumentum now face investor scrutiny over the economics and execution of their proposed combination. Anthropic’s IPO process will put its revenue trajectory alongside its compute obligations, while OpenAI’s Abu Dhabi agreements and national-security hire will clarify how its government and infrastructure work develops. The Brave lawsuit and Anthropic’s Claude Code restrictions remain unresolved questions about the boundaries of AI access to existing content and software.