Government & Policy | 2 min read

Former FTC Chair Lina Khan Calls AI Industry 'Constitution' a Proven Failure

Lina Khan told ABC News the voluntary AI safety 'constitution' signed by tech CEOs is a replay of a self-regulation playbook that already failed once — and calls for applying existing antitrust and consumer protection law directly to AI.

Hector Herrera
Hector Herrera
A government building interior featuring data centers, related to Former FTC Chair Lina Khan Calls AI Industry 'Constitution'  from an unusual angle or perspective
Why this matters Lina Khan told ABC News the voluntary AI safety 'constitution' signed by tech CEOs is a replay of a self-regulation playbook that already failed once — and calls for applying existing antitrust and consumer protection law directly to AI.

Former Federal Trade Commission Chair Lina Khan went on national television Sunday to call the AI industry's new voluntary safety "constitution" a repeat of a playbook that already failed — and to warn lawmakers not to make the same mistake twice.

Appearing on ABC's This Week, Khan argued that the voluntary commitments signed by major AI company CEOs this week mirror the self-regulatory promises the tech industry made roughly a decade ago around children's privacy, content moderation, and competitive practices. Those promises, she said, produced documented harms to children and unprecedented market concentration — and extending the same deference to AI would be "an enormous mistake."

Khan's Argument

Her case has three parts:

1. Self-regulation doesn't hold. Big tech made similar voluntary commitments before. The result was years of inaction while harms accumulated, followed by congressional hearings that produced little legislation. Khan's implicit point: the window to act before AI is deeply embedded in critical systems is narrowing.

2. The tools already exist. Khan isn't calling for new AI-specific legislation. She's calling for the aggressive application of existing consumer protection and antitrust law to AI products and the companies behind them. The FTC has authority over unfair or deceptive practices; AI systems that discriminate, manipulate, or deceive fall squarely within that mandate — if the agency chooses to use it.

3. Voluntary is not accountable. A "constitution" signed by CEOs carries no legal weight. There's no enforcement mechanism, no independent audit requirement, and no penalty for non-compliance. Companies can sign and ignore.

Why It Matters Now

Khan's appearance comes at a specific moment: the Trump White House is simultaneously creating a new federal AI task force and signaling openness to industry self-governance over new regulation. The AI safety "constitution" fits neatly into that framework — industry shows it's serious, administration avoids binding rules.

Khan, who clashed repeatedly with major tech companies during her FTC tenure and was replaced at the start of the Trump administration, represents the opposing view: that market-distorting concentration and consumer harm don't wait for voluntary fixes.

What to Watch

Whether her argument gains traction depends on Congress. Democratic members have used Khan's framework before; whether Republicans with AI constituents in their states — semiconductor manufacturers, data centers, AI startups — engage seriously with her enforcement-first argument will shape the policy trajectory heading into 2027.

Source: ABC News, October 4, 2026

Key Takeaways

  • ✓ 1. Self-regulation doesn't hold.
  • ✓ 2. The tools already exist.
  • ✓ 3. Voluntary is not accountable.

Did this help you understand AI better?

Your feedback helps us write more useful content.

Hector Herrera

Written by

Hector Herrera

Hector Herrera is an AI systems architect in Houston and founder of Hex AI Systems. He designs and runs AI systems in production and writes daily about how AI is reshaping business, government and everyday life. 20+ years building for the web. Houston, TX.

More from Hector →

Get tomorrow's AI briefing

Join readers who start their day with NexChron. Free, daily, no spam.

More from NexChron