“If you don’t have control over AI and it’s integrated into the military, integrated into the national security state, then our civilian and military decision makers and policymakers would then not be able to control the most powerful technology on the planet and the most powerful weapons on the planet,” Steinhauser explained.
“It could potentially create social and political instability and lead to really awful things happening because people are, so many people are out of work, they don’t have anything to do,” Steinhauser explained.
“The clear thing to do here is to have meaningful safeguards on AI,” Steinhauser said.
The White House released legislative recommendations for artificial intelligence (AI) on Friday morning and said the Trump administration will work with Congress in the coming months to pass legislation.
The national framework aims to address six objectives to ensure Americans trust the budding technology, but an AI expert believes the administration is not addressing the top concerns that deal with national security and job loss.
The six objectives the administration is recommending Congress address include:
- Protecting children and empowering parents
- Safeguarding and strengthening American communities
- Respecting intellectual property rights and supporting creators
- Preventing censorship and protecting free speech
- Enabling innovation and ensuring American AI dominance
- Educating Americans and developing an AI-ready workforce
The White House said the framework will only succeed if it is applied uniformly across all 50 states. The framework reads, “Congress should preempt state AI laws that impose undue burdens to ensure a minimally burdensome national standard consistent with these recommendations, not fifty discordant ones.”
The Texas legislature did pass AI regulations during the 89th legislative session last year. The regulations prohibited the use of AI to create a social scoring system, produce and distribute child pornography, and gave enforcement powers to the Texas Attorney General, among other regulations.
Brendan Steinhauser is the CEO of Alliance for Secure AI Action, a nonprofit that educates the public about the implications of advanced AI. Steinhauser said he does not believe the national framework presented by the White House addresses his top concerns.
His chief concern is a scenario where humans lose control of artificial superintelligence (ASI), a more advanced form of AI. He’s talking about something known as the alignment problem, the idea that we are unable to ensure that ASI will align and operate with human values and ethics.
“If you don’t have control over AI and it’s integrated into the military, integrated into the national security state, then our civilian and military decision makers and policymakers would then not be able to control the most powerful technology on the planet and the most powerful weapons on the planet,” Steinhauser explained.
The national framework does say that Congress should ensure that national security agencies establish plans to “mitigate potential concerns.”
Another concern for Steinhauser that he feels is not addressed in the White House’s release is job loss. He references AI CEOs who have said publicly that AI will replace human workers. The predictions from AI leaders have varied, but Elon Musk, for example, said he believes AI and robots will replace all jobs and that working will be optional. Steinhauser believes replacing human workers will have detrimental impacts on society.
“It could potentially create social and political instability and lead to really awful things happening because people are, so many people are out of work, they don’t have anything to do,” Steinhauser explained.
He believes the country should focus more on narrow uses of AI, where humans integrate it into their work and daily lives, rather than a race for superintelligence.
“The clear thing to do here is to have meaningful safeguards on AI,” Steinhauser said.