Impact of the AI Whistleblower Protection Act on Tech Employees

On May 15, the AI Whistleblower Protection Act (AIWPA) was introduced by Senate Judiciary Chairman Chuck Grassley (R-IA), signaling a bipartisan effort to shield individuals who expose potential artificial intelligence security flaws or breaches. This bill safeguards current and former employees, as well as independent contractors, who disclose information in good faith, without requiring them to prove a violation of the law.

The introduction of the AIWPA is crucial in the current landscape of AI safety, as regulations are needed to establish transparency and accountability standards in the public interest. Given the possibility of increased deregulation at both state and federal levels, protections for whistleblowers are essential to hold the AI industry accountable.

Recent developments in Washington highlight the push for minimal regulation in the AI sector. Tech industry leaders advocated for “light touch” regulations during a Senate hearing on the global AI competition, while the House included a provision in a budget reconciliation bill to prevent states from regulating AI with a moratorium. In an under-regulated environment, whistleblower disclosures play a vital role in aiding law enforcement investigations, as demonstrated by various government agency whistleblower programs.

Whistleblowing has been a long-standing focus for Grassley, particularly evident in his efforts to strengthen the False Claims Act in 1986. Recently, his office received a letter from whistleblowers concerning alleged illegal agreements at OpenAI, prompting Grassley’s commitment to whistleblowers as essential watchdogs in the rapidly evolving AI industry.

The AIWPA defines an AI security vulnerability as any breach that could result in unauthorized access to a company’s technology, while an AI violation encompasses any federal law breach linked to AI development, deployment, or use. The bill also outlines a broad scope for reporting disclosures to federal agencies, the attorney general, Congress, or internal compliance programs within companies.

In cases of retaliation, AI whistleblowers are protected under the AIWPA and stand to gain remedies such as job restoration, back wages, and damages through federal courts. The legislation prohibits employers from retaliating against whistleblowers for their protected activities and ensures that employment contracts or forced arbitration clauses cannot undermine these protections.

While the AIWPA offers crucial safeguards, advocates argue for an expanded definition of AI violations. Including failures to adhere to internal safety protocols in the bill could address employee concerns, such as rushed testing protocols at OpenAI, enhancing the bill’s ability to capture misconduct impacting the public interest.

By spotlighting the significance of whistleblowing, the AIWPA could alleviate apprehensions within the AI industry and empower employees to exercise their rights to a greater extent. While the bill lacks provisions for monetary rewards, industry-wide standards set by Congress may eventually offer such incentives, a step yet to be taken in the AI sector.