AI News: Netflix vs ByteDance, Meta AI Lobbying, Anthropic Crisis

This week's AI news cycle is dominated by power plays — studios fighting back against generative AI piracy, Meta spending tens of millions to shape AI legislation, and the US Department of Defense potentially blacklisting one of Silicon Valley's most prominent AI companies. Here's what B2B leaders need to know.

Netflix Threatens ByteDance With "Immediate Litigation" Over Seedance AI

Netflix has given ByteDance three days to respond to a cease-and-desist letter over its Seedance 2.0 AI video generation tool, threatening "immediate litigation" if the company doesn't act.

The streaming giant joins Disney, Paramount, and Warner Bros. in accusing ByteDance of enabling mass copyright infringement. Netflix's letter — signed by litigation director Mindy LeMoine — calls Seedance "a high-speed piracy engine, generating mass quantities of unauthorized derivative works utilizing Netflix's iconic characters, worlds, and scripted narratives."

The specific allegations are striking. Users have generated unauthorised clips featuring Stranger Things characters and monsters, Bridgerton Season 4 costumes (including Sophie Baek's "Lady in Silver" gown), Squid Game environments with real-world figures inserted, and KPop Demon Hunters character reproductions. ByteDance reportedly promoted some of this content using #Bridgerton tags through its own official social media channels.

ByteDance announced additional guardrails on Monday, but Netflix and Warner Bros. both sent letters on Tuesday regardless. Netflix's letter notably preempts a fair use defence, arguing that "use of copyrighted works to create a competing commercial product, especially one that regurgitates the original, is not protected by fair use."

What this means for B2B: The Seedance controversy accelerates the legal reckoning around AI-generated content. Any company using generative AI for content creation — marketing materials, product demos, sales collateral — needs clear provenance tracking and licensing for training data. The "move fast and apologise later" era of AI content generation is ending.

Meta Plans $65 Million Political Spend to Shape AI Legislation

Meta is planning to spend $65 million on US elections to influence AI-related legislation, according to the New York Times. The funding will flow to pro-AI super PACs, including two new ones: the Republican-focused "Forge the Future Project" and the Democrat-focused "Making Our Tomorrow."

The PACs will back politicians who support AI-friendly policies and push back against legislation that could limit Meta's AI business growth.

This bipartisan approach signals how seriously Big Tech takes the current regulatory environment. With the EU AI Act already in effect and US states passing their own AI legislation at an increasing pace, Meta is making a calculated bet that shaping the rules is cheaper than complying with unfavourable ones.

What this means for B2B: The regulatory landscape for AI is being actively shaped by well-funded lobbying. European companies operating under the EU AI Act already have a compliance framework — the question is whether US regulation will converge toward similar standards or diverge entirely. Companies building AI systems should plan for the stricter standard regardless.

Anthropic Launches Claude Sonnet 4.6 — Approaches Opus-Level Intelligence

Anthropic released Claude Sonnet 4.6, calling it their "most capable Sonnet model yet." The upgrade brings significant improvements across coding, computer use, long-context reasoning, agent planning, and knowledge work — with developers in early access often preferring it to Claude Opus 4.5, Anthropic's most powerful model from November 2025.

The standout feature is computer use capability. Anthropic's progress on the OSWorld benchmark — which tests AI on real software tasks across Chrome, LibreOffice, VS Code, and more — shows steady gains over sixteen months. Early users report "human-level capability" in tasks like navigating complex spreadsheets and filling out multi-step web forms.

Pricing remains unchanged from Sonnet 4.5 at $3/$15 per million input/output tokens, with a 1M token context window now in beta. It's the new default for free and pro Claude users.

What this means for B2B: Computer use is the frontier that matters most for enterprise automation. When AI models can reliably operate legacy software — the ERP systems, CRMs, and specialised tools that don't have modern APIs — the automation opportunity expands dramatically. Sonnet 4.6's improvements suggest we're closer to practical, production-grade computer-using AI agents than many realise.

US Department of Defense May Blacklist Anthropic as "Supply Chain Risk"

In a remarkable twist, the same week Anthropic launched its most capable model, Axios reported that the Department of Defense is considering designating Anthropic as a "supply chain risk." If enacted, anyone wanting to do business with the US military would need to cut ties with the company entirely.

The two sides have reportedly been negotiating for months over how the military can use Anthropic's AI tools. The designation would effectively force defence contractors and government agencies to choose between Anthropic and military contracts.

This comes amid broader tensions between the safety-focused AI labs and the current administration's push for rapid AI deployment in defence applications.

What this means for B2B: Vendor risk in AI is no longer just about uptime and pricing — it's geopolitical. Companies building on any single AI provider's technology should have multi-model strategies and contingency plans. If the DoD can designate a leading AI company as a supply chain risk, any customer dependency on a single provider carries material business risk.

xAI Faces NAACP Lawsuit Over Unpermitted Data Centre Pollution

The NAACP has sent a notice of intent to sue Elon Musk's xAI, accusing the company of illegally installing gas turbines in Mississippi to power its Colossus 2 data centre without permits. Thermal drone images revealed more than a dozen turbines operating at the site, according to a Guardian/Floodlight investigation.

This isn't xAI's first data centre controversy. The company has faced repeated criticism for prioritising speed of AI infrastructure buildout over environmental compliance and community impact.

What this means for B2B: The environmental cost of AI infrastructure is becoming a material business concern. European companies evaluating AI providers should consider the sustainability credentials of their supply chain — especially as ESG reporting requirements tighten under the Corporate Sustainability Reporting Directive (CSRD). Choosing providers with transparent, compliant infrastructure isn't just ethics — it's risk management.

The Week Ahead

The Seedance litigation timeline gives ByteDance until roughly February 21st to respond — expect escalation if they don't. Meanwhile, Anthropic's dual situation (best product launch alongside potential government blacklisting) perfectly illustrates the contradictions in the current AI landscape: the technology is advancing faster than the institutions trying to govern it.

For European B2B companies, the takeaway is consistent: build on multi-vendor architectures, maintain strict content provenance, and treat regulatory compliance as competitive advantage rather than overhead. The companies that get this right now won't need to scramble later.

Want to build an AI strategy that's resilient to vendor risk and regulatory shifts? Book a 30-minute strategy call to assess your current AI architecture and identify your biggest exposure points.