Earlier this month, OpenAI whistleblowers submitted a complaint with the Securities and Exchange Commission (SEC) claiming the company’s non-disclosure confidentiality agreements (NDAs) prevent them from voicing concerns about the irresponsible advancement of artificial intelligence (AI) technologies. They also contend that, considering AI companies hold extensive confidential information about the strengths and weaknesses of their systems, technology employees should feel free to raise concerns to federal regulatory authorities.  

The whistleblowers appear to suggest that the firms are prioritizing profit over safety in creating their technology and are asking the SEC to approve an investigation of OpenAI’s previous NDAs, assess the ongoing efforts by the company to adhere to SEC regulations, and urge AI companies to foster workplace cultures that support employees in voicing concerns. But regardless of what the SEC decides to do, Congress will not be far behind with regulation. 

In fact, it may be inevitable that Congress will act to regulate AI – and when it does, it should have the most accurate information possible. Meanwhile, the tech sector’s seeming reluctance to disclose AI risks may amplify concerns and restrict Congress’ ability to craft effective legislation that mitigates potential harms without unnecessarily hindering AI development. 

Before Congress acts, it will almost certainly require a little tech know-how from AI firms. AI systems are complex, and as the technology booms, it brings with it concerns about potential biases, data privacy violations, and misuse – and recent news only intensifies these fears. Other conversations about AI have raised fears about workers being replaced, of a lack of understanding of how AI systems operate, and the uncertainty they pose for humanity. It is not likely that Congress will be able to grasp all of these nuances without significant industry collaboration. With such cooperation, however, a responsible AI framework could be crafted to include transparency measures to ensure the public can generally understand how systems work and are developed, as well as safety measures to ensure they do not negatively impact society. 

Regular and intentional collaboration between Congress and AI developers would ensure that whatever regulations are eventually passed are at least targeted, clear, and useful. In the end, effective AI regulation will require the joint efforts of policymakers and the tech industry to shed light on the benefits and risks of AI.