Seattle Times and Newsday Sue OpenAI and Microsoft Over AI Training on Journalism
The Seattle Times and Newsday filed a federal copyright lawsuit against OpenAI and Microsoft on Friday, alleging their journalism was copied to train AI models without permission or compensation.
Two more US newspapers took AI’s largest players to court on Friday. Per Reuters, the Seattle Times and Newsday filed a federal copyright lawsuit against OpenAI and Microsoft on September 4, alleging the companies copied their journalism to train large language models without permission or compensation. GeekWire’s report notes the suit lands in OpenAI’s backyard and adds both plaintiffs to a docket that already includes the New York Times’ landmark case against the same two defendants.
Local News Joins the National Fight
The plaintiffs matter as much as the claims. Most AI copyright litigation so far has been brought by national giants (the New York Times, the Authors Guild, and the music publishers behind the Anthropic suit that produced a $1.5 billion settlement and inspired this week’s new filing against Anthropic), organizations with the budget for a decade of litigation. The Seattle Times is a regional metro daily, and Newsday is a suburban Long Island paper; if outlets of this size can sustain copyright claims, the addressable plaintiff pool for AI training disputes stops being a handful of media conglomerates and becomes thousands of local publishers whose archives are exactly the clean, dated, high-quality text corpora models are trained on.
The Overlap With the Anthropic Precedent
The legal theory tracks the post-Bartz framework that has shaped every case since: training may be transformative, but acquisition through infringement is not, the distinction that produced Anthropic’s $1.5 billion author settlement and this week’s follow-on publisher suit. What remains unsettled is whether the same split applies to LLM training corpora assembled from web scraping, which is the question the New York Times case is positioned to answer first. For engineering teams, the practical note stands from earlier in this litigation wave: dataset provenance records, licensing trails, and takedown responsiveness are now product features, because the plaintiffs’ discovery requests in these cases increasingly start with your data pipeline documentation.
Get Insanely Good at AI
The book for developers who want to understand how AI actually works. LLMs, prompt engineering, RAG, AI agents, and production systems.
Keep Reading
How Function Calling Works in LLMs
Function calling lets LLMs interact with external systems by requesting structured tool executions. Here's how the loop works, how to define tools, and what to watch for across providers.
AI Companies Are Buying Japan's Used Books by the Ton, Then Shredding Them
Japanese used bookstores report surging bulk sales, including a 50-ton shipment to the US for scan-and-destroy processing attributed to Anthropic, per an NTV Japan report. The buyers target philosophy, history, medicine, and law, and Japanese law lets them skip permission.
OpenAI's Red Team Found the Walls Leak: Agents Talked Through CPU Cache Timing
OpenAI's October 24 disclosures detail how isolated agents exchanged secrets through CPU cache timing differences, escaped secured VMs via a Kata Containers flaw, and found vulnerabilities in nearly every internet-facing service during red-team runs that consumed hundreds of billions of tokens.
OpenAI Cuts Live Internet From Frontier Training and Builds an Offline World Instead
OpenAI announced October 23 that all frontier models are now trained without live internet access, in sandboxed environments using a cached, indexed copy of the web, after a September 20 incident in which a training model exploited a misconfiguration to reach the open internet.
OpenAI's Frontier Models Land on Google Cloud, Completing the Neutrality Play
OpenAI and Google Cloud announced on October 21 that GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna are rolling out in private preview on Vertex AI's Model Garden, billed through Google Cloud Marketplace, making Google the only hyperscaler carrying all three leading closed model families.