Comparing OpenAI, Anthropic, Google AI, and Microsoft Copilot
OpenAI, Anthropic, Google AI (including DeepMind and Gemini), and Microsoft Copilot dominate AI development. Recent reports highlight their regulatory pressures, model behaviors, and business standing. All face growing oversight, while their tools show similar effects on human tasks.
Shared Regulatory Demands
California’s SB 53 and New York’s RAISE Act hit OpenAI, Anthropic, and Google directly. These laws force the companies to add whistleblower protections for staff, report major safety events to state offices, and explain model testing for risks. NBC News details this in coverage of a White House AI framework, which pushes for federal rules over state ones to avoid a “patchwork.” The framework also seeks to cap developer liability and block states from punishing AI firms for third-party misuse.
Anthropic ran into extra trouble: the Trump administration cut it off from government contracts, calling it “woke.” Anthropic sued, claiming First Amendment violations.
Microsoft Copilot, baked into its ecosystem, dodges specific mentions in these state laws but operates under the same federal push.
Model Behavior in Real Tests
A study from University of Washington researchers tested three models: Anthropic’s Claude 3.5 Haiku, OpenAI’s GPT-5 Mini, and Google’s Gemini 2.5 Flash. Heavy users generated essays on money and happiness that turned neutral 69% more often than light or no-AI users. Outputs got formal, dropped pronouns by 50%, and lost personal voice. Participants felt their work was less creative but stayed satisfied anyway.
Natasha Jaques, a Google DeepMind researcher and study lead, said LLMs fail to match human style. They rewrite substance. NBC News reported the peer-reviewed findings, accepted for an AI conference workshop. Human editors tweak words; these AIs overhaul text, shifting meaning.
Microsoft Copilot wasn’t in the test, but its integration across Office tools raises parallel questions about everyday writing shifts.
Business and Market Snapshot
- OpenAI: Powers ChatGPT lineage like GPT-5 Mini; heavy regulation target. Eightco added $40 million to its stake for a $90 million total investment, Morningstar reports.
- Anthropic: Claude leader; gov contract loss, now in court.
- Google AI: Gemini and DeepMind drive research; state-law compliant.
- Microsoft Copilot: Ties to strong stock performance, listed at $269 CMP in March 2026 by Analytics Insight among top AI plays with NVIDIA and Broadcom, which The Motley Fool calls unstoppable buys.
Apple hit $900 million in AI revenue without leading AI tech, Gadget Hacks reports.
These companies fuel AI growth amid rules and scrutiny. OpenAI, Anthropic, and Google shoulder early compliance costs, while Microsoft rides infrastructure waves.
