The groups called on the government to mandate companies to report energy use and emissions throughout the AI life cycle and provide public explanations on how models produce information. They also requested a government study on AI's effect on the spread of climate disinformation and the implementation of safeguards against mass production of disinformation. The Biden executive order currently instructs various government agencies to develop technical reporting standards for large AI developers.
Key takeaways:
- Several groups have criticized the Biden administration's approach to AI regulation, stating it does not adequately address AI's impact on climate change and its potential use in spreading disinformation.
- The groups, which include Accountable Tech, Amazon Employees for Climate Justice, and Greenpeace USA, have sent a letter to the White House urging for more climate-focused policies in the AI executive order.
- The letter accuses the administration of neglecting the huge energy cost of training large language models (LLMs) and their potential use in spreading climate disinformation. It calls for companies to report energy use and emissions throughout the AI lifecycle and provide public explanations on how models produce information.
- The Biden executive order does mention climate change, directing government agencies to explore the use of AI in combating it. However, the groups argue that more needs to be done, particularly in studying the effect of AI on the spread of climate disinformation and implementing safeguards against it.