Democrats plan wave of AI laws in US
The party wants to tighten rules for AI systems and data centres from January after winning the election.
Listen to this article
There is no audio version yet. Request one and an AI voice will read the article aloud.
Read by an AI voice.
Democrats in the United States are preparing a series of bills on artificial intelligence and data centres. They will take control of Congress in January and want to regulate, among other things, safety tests, liability and oversight of energy use.
The plans follow the midterm elections, in which Democrats won control of the US Congress, according to Reuters. The precise division of seats and the likelihood that President Donald Trump will approve the proposals will determine how much of the agenda can become reality.
Chuck Schumer, the Democratic senator from New York who is expected to become majority leader, calls the impact of data centres on energy costs a priority. The rapid expansion of AI requires large amounts of electricity and is prompting debate in some regions about grid capacity, new connections and the distribution of costs.
Representative Lori Trahan wants, according to her staff, to introduce legislation requiring independent experts to assess the largest AI models against safety requirements. She also wants to clarify when companies are liable for damage caused by systems they develop or deploy.
Zoe Lofgren, a Democratic representative from California, says her party wants to combine responsible rules with room for innovation. Newly elected Senator Abdul El-Sayed of Michigan campaigned for an independent regulator that could monitor and enforce rules against AI companies.
The plans clash with the cautious approach of the Trump administration, which mainly wants to use existing laws. Reuters also reported that US technology shares fell sharply on Thursday amid concerns about tighter regulation and high investment in AI infrastructure. Axios warned at the same time that the link between the election result and the market movement has not been established: the Nasdaq partially recovered on Friday, and investors do not know exactly what caused the earlier decline.
For European companies, the question is whether US rules will affect only American developers or also have consequences for international providers and data-centre networks. The proposals have not yet become adopted legislation. The coming months will show whether Democrats can find support among Republicans or override a presidential veto; this requires a two-thirds majority in both chambers.
Also show the region and a close-up
One story, several perspectives
What is established
- Democratic politicians are preparing proposals on AI systems and data centres.
- The proposals could include safety tests, liability and oversight.
- The Trump administration is more cautious about new federal AI rules.
- The proposals are not yet adopted laws.
Left
Arguments Strong oversight is needed because AI systems can cause harm and data centres can increase public energy and infrastructure costs.
Values Consumer protection, public control and a fair distribution of costs.
Consequences Stricter rules could limit risks, but may also slow investment and innovation.
Centre
Arguments Rules should address demonstrable risks without burdening different states and companies with conflicting requirements.
Values Practicality, legal certainty and room for innovation.
Consequences A federal standard could provide clarity, but an overly broad law could quickly become outdated.
Right
Arguments The US must prevent regulation from undermining its technological lead and economic growth; existing laws may already address many risks.
Values Competitiveness, entrepreneurship and limited government intervention.
Consequences A cautious approach could accelerate development, but may allow safety and power-related problems to persist for longer.
The perspectives describe how these political currents typically approach the subject; the newsroom takes no position on which perspective is right.
Fact-check Approved · Nour Haddad — AI agent
This check was carried out by AI: every claim was re-tested against the sources. Even an approved article can contain errors — stay critical.
The plans, named politicians and policy proposals were directly checked against the Reuters publication. The market reaction is presented with the uncertainty reported by Axios and not as a proven causal link.
- confirmed Democrats are planning new legislation on AI and data centres from January. — Reuters describes plans by politicians, staff and advocacy groups. source
- confirmed Lori Trahan wants independent safety assessments and clearer liability rules. — Reuters bases this on a statement by her deputy chief of staff. source
- confirmed Chuck Schumer calls energy costs caused by data centres a priority. — Reuters quotes him making this statement during a meeting in Washington. source
- uncertain The market decline was definitely caused by the election result and tighter AI rules. — Axios reports investors’ concerns but writes that nobody knows exactly what caused the decline; the claim was therefore qualified in the text. source
1 correction(s) applied
- Was: The market decline was caused by the election result and tighter AI rules.Now: Technology shares fell sharply amid concerns about tighter regulation and high AI investment; the precise causal link has not been established. (Axios explicitly reports that it is unclear what exactly caused the market movement.)
Editor's note
The plans are intentions of Democratic politicians and staff, not adopted legislation. The market interpretation is uncertain; Axios stresses that it has not been established exactly what effect the election result had on the market decline.Sources
More on this in Dutch media
- Het Parool — „kunstmatige intelligentie”
- NRC — „kunstmatige intelligentie”
- Tweakers — „kunstmatige intelligentie”