GenAI Daily - February 21, 2026: ByteDance Seedance 2.0 Draws Hollywood Backlash, Anthropic Blocks Third-Party OAuth, OpenAI Joins UK AI Safety Initiative

Top Story

ByteDance's Seedance 2.0, launched this week as one of the most sophisticated AI video generation models to date, has triggered fierce resistance from Hollywood studios and creators who say the tool enables "blatant" copyright infringement. The model generates 15-second videos from text prompts using images, audio, video and text simultaneously, producing content at dramatically lower costs than traditional video production.

The controversy erupted after viral demonstrations showed Seedance creating videos of Tom Cruise fighting Brad Pitt from a simple two-line prompt, prompting "Deadpool" screenwriter Rhett Reese to declare "It's likely over for us." The Motion Picture Association quickly demanded ByteDance "immediately cease its infringing activity," while Disney and Paramount have issued cease-and-desist letters alleging widespread IP theft.

ByteDance has responded by implementing better safeguards and rolling back features like voice cloning based on images alone.

Why it matters: The backlash reveals the collision between rapidly advancing AI capabilities and established creative industries. Unlike gradual AI adoption, video generation tools can immediately replicate copyrighted characters and styles, forcing legal battles that will shape how generative AI navigates intellectual property at scale.

Key Developments

Anthropic Bans Third-Party OAuth Access for Claude Code

Anthropic has updated its legal documentation to explicitly prohibit using OAuth tokens from Free, Pro, and Max subscription plans in third-party tools or services, including its own Agent SDK. The policy change disrupts developers who had been using subscription credentials with popular coding tools like OpenCode and OpenClaw.

The company directs businesses to use API keys with usage-based billing instead of subscription pass-through authentication. Anthropic reserves enforcement rights without prior notice, having already blocked Max OAuth access in third-party clients starting January 9, 2026.

Source

Impact: Forces developers to migrate from flat-rate subscriptions to pay-per-token billing, fundamentally changing cost structures for AI coding workflows and highlighting platform control dynamics.

OpenAI and Microsoft Join UK AI Safety Coalition

OpenAI and Microsoft have joined the UK's AI Security Institute (AISI) Alignment Project, pledging £5.6 million and additional support to bring total funding to £27 million. The initiative aims to ensure advanced AI systems remain safe, secure and under human control.

The first round has awarded grants to 60 research projects across 8 countries, with a second round opening this summer. AI alignment research focuses on making sure AI models operate as intended even as their capabilities rapidly evolve, addressing what experts consider one of the most urgent technical challenges.

Source

Impact: Signals industry recognition that AI safety research requires coordinated international funding as systems become more autonomous and capable of complex reasoning.

Product Launches

• OpenAI Frontier:

Enterprise platform allowing technical and non-technical teams to hire AI coworkers that complete tasks using files, code, and tools in a dependable execution environment. Runs across local, cloud, and OpenAI-hosted runtimes without forcing workflow changes.

(OpenAI)

• Y Combinator W2026 Batch:

Includes enterprise AI startups like Crow (lets users control apps through chat with API integration) and Jinba (enables teams to "vibe-code" AI workflows in plain language without requiring engineers).

(Y Combinator)

Funding & Deals

• ElevenLabs - $500M Series D:

Voice AI company raised funding led by Sequoia at an $11 billion valuation, targeting realistic voice synthesis and audio generation applications.

• Led by Sequoia (TechCrunch)

• Runway - $315M Series E:

Media-generation platform valued at $5.3 billion with funding led by General Atlantic, with participation from Nvidia, Fidelity, and Felicis for video creation tools.

• Led by General Atlantic (TechCrunch)

• Fundamental - $255M Series A:

AI research company reached $1.4 billion valuation with investors including Oak HC/FT, Salesforce Ventures, Valor Equity Partners, and QP Ventures.

• Multiple investors (TechCrunch)

Tomorrow's Watch List

• Second round applications for UK AI Safety Institute Alignment Project open this summer
• ByteDance expected to announce enhanced IP protection measures for Seedance 2.0
• Additional enterprise AI agent platforms likely to launch following OpenAI Frontier success

Related reading: Check out this week's [Deep Insights analysis] for strategic context on AI platform control and enterprise adoption patterns.