The Trump administration released its highly planned AI action plan on Wednesday. This has stripped a sharp shift from former President Biden’s careful approach to addressing AI risks, instead building AI infrastructure, cutting red tape for high-tech companies, strengthening national security and moving forward with plans to compete with China.
This downstream impact of this shift is likely to ripple across a variety of industries and can be felt by the average American consumer. For example, AI Action Plans downplay efforts to mitigate possible harms from AI, and instead prioritize building data centers to power the AI industry, even if they mean using federal land or providing power during critical energy grids.
However, much of its effect depends on how the AI action plan is implemented, and many of these details have not yet been sorted out. AI Action Plans are more of an action blueprint than a step-by-step instruction book. But the direction is clear. Progress is the king.
The Trump administration has positioned it as the only way to “hear the arrival of a new golden age of human prosperity, the goal is to convince Americans that it is their greatest benefit to spend billions of dollars on building data centers. Part of the plan also includes policy proposals to expand workers and partner with local governments to create jobs related to work in the data center.
“We must utilize the full power of American innovation to ensure our future,” Trump said in a statement. “To do this, we will continue to reject the fundamental climate doctrine and bureaucratic deficits, as the administration has done since inauguration. Simply put, we need ‘Build, Baby, Build!”
The AI Action Plan is written by the Trump administration’s technology team and AI experts, many of which come from Silicon Valley companies. This includes the office of Michael Kratsios, Director of Science and Technology Policy. AI and Crypto Czar David Sacks; assistant to national security president Marco Rubio. Over 10,000 interest groups submitted public comments considered in the plan.
Deregulation and AI moratorium take-home
Earlier this month, the Senate removed the controversial provisions in the budget bill that would ban states from regulating AI for 10 years. That provision would tie the state’s federal broadband funding to comply with the suspension if it was included in the bill.
TechCrunch Events
San Francisco
|
October 27th-29th, 2025
The issues seem to be on the rise as AI action plans explore new ways to prevent AI adjustments. As part of a broader mission to “unleash prosperity through deregulation,” the administration is threatening to limit state federal funds under AI regulations.
The plan also directs the Federal Communications Commission to “assess whether state AI regulations interfere with their obligations and the ability of agencies to implement their agencies.” In other words, if state AI regulations are exposed to radio, television and the internet (many of them do), the FCC could be involved.
At the federal level, the Action Plan instructs the Office of Science and Technology Policy to ask businesses and the public about current federal regulations that would hinder AI innovation and adoption so that federal agencies can take appropriate action.
Cutting the deficit around the data center
Trump’s call for deregulation extends to how the administration wants to encourage the creation of AI-related infrastructures, including data centers, semiconductor fabs and power supplies. The administration argues that existing environmental regulations, such as the NEPA, the Clean Air Act and the Clean Water Act, are hampering America’s needs to meet the rapid demands of the AI arms race.
That’s why Trump’s AI Action Plan focuses on stabilizing the US energy grid. At the same time, the plan calls on the federal government to find new ways to enable large consumers, such as AI companies, to manage their power consumption during critical grid periods.
Certain companies, such as Xai and Meta, have been criticized for focusing pollution on vulnerable communities. Critics accused Xai of bypassing environmental protection measures and using Memphis data centers to expose residents to harmful emissions from gas-powered turbines.
The Action Plan calls for creating category exclusions, streamlining process permissions, and expanding the use of fast track programs such as FAST-41 to facilitate critical AI infrastructure in federal lands, including national parks, federally protected wilderness areas and military bases, among other things.
Returning to other Trump themes that beat China, the strategy focuses on highlighting the lockout of foreign technology and security protections to maintain “hostile technologies” such as Chinese-made chips and hardware from the US supply chain.
Trump’s war on “biased AI”
One of the main standouts of Trump’s AI Action Plan focuses on protecting freedom of speech and “American values” by eliminating misinformation, DEI, and references to climate change from the federal risk assessment framework.
“These systems are built from scratch with freedom of speech and expression in mind, and it is essential that US government policies do not hinder their objectives,” the plan reads. “We must ensure that freedom of speech flourishes in the age of AI and that the AI sourced by the federal government objectively reflects the truth, not the agenda of social engineering.”
Despite the intention to ensure that government policies do not interfere with free speech, AI action plans could do just that.
One recommended policy action is to update federal procurement guidelines to ensure that governments only contract with large frontier language model developers who ensure that the system is objective and free from top-down ideological bias.
The language is similar to that reported in Trump’s executive order reported by the Wall Street Journal, and is expected to be released later today.
The problem is that objectiveness is difficult to achieve and the government has not yet defined a plan to evaluate models based on neutrality.
“The only way to be neutral is literally unengaged,” Rumman Chowdhury, data scientist and CEO of Tech Nonprofit Human Intelligence, told TechCrunch that he is the former National Science Envoy for AI.
Humanity, Xai, Google and Openai have secured all government contracts worth up to $200 million each to integrate AI applications into the Department of Defense. The implications of Trump’s policy proposals and his impending executive orders could be far-reaching.
“The order, for example, that “companies that produce neutral AI models, such as AI models, are likely to violate the initial amendment,” said Eugene Volok, an American legal scholar specializing in issues of the First and Second Amendments, in an email. “An order that says, “We will enter into a contract to buy a fully neutral model” is more constitutionally defensible, but of course it may be very difficult to implement effectively (partly because it is difficult to know what “neutral” is in these situations). ”
He added: “If this command instructs the agency to select AI based on a combination of accuracy and neutrality, leaving each agency with some latitude to determine what it means, that might be more feasible.”
Encourage an open approach to AI
Trump’s AI Action Plan aims to encourage the development and adoption of open AI models. It is available for free download online and is created with American values in mind. This appears to be a response primarily to the rise of open AI models from Chinese AI labs, including Deepseek and Alibaba’s Qwen.
As part of his plan, Trump wants to ensure that startups and researchers working on open models have access to large computing clusters. These resources were expensive and were usually only possible for high-tech companies that could attack millions or billion dollar contracts with cloud providers.
Trump also hopes to partner with leading AI model developers to increase access to private AI models and data in the research community.
American AI companies and organizations that have taken an open approach that includes meta, AI2 and embracing faces can benefit from Trump’s open AI embrace.
AI Safety and Security
Trump’s AI Action Plan includes several provisions to meet the AI safety community. One of these efforts includes the launch of a federal technology development program to study AI interpretability, AI control systems, and adversarial robustness.
Trump’s plan also directs federal agencies, including the Department of Defense and the Department of Energy, to host hackathons to test AI systems for security vulnerabilities.
Trump’s plans also acknowledge the risks of AI systems that contribute to cyberattacks, as well as the development of chemical and biological weapons. The plan asks frontier AI model developers to work with federal agencies to assess these risks and asks how they will put American national security at risk.
Compared to Biden’s AI executive order, Trump’s plans have less focused on requiring major AI model developers to report safety and security standards. Many tech companies argue that reporting safety and security is a “stingy” task and that Trump seems to want to limit it.
China’s restrictions
Perhaps unsurprisingly, Trump is bringing war with China into AI competition with his plan of action. Most of Trump’s AI action plans focus on preventing the threat of national security from accessing advanced AI technologies.
Under Trump’s plan, federal agencies will work together to gather information on foreign frontier AI projects that could threaten US national security. In one of these efforts, the Commerce Department is tasked with evaluating China’s AI model to match the talk points and censorship of China’s Communist Party.
These groups will also conduct assessments on the level of AI adoption among American enemies.
National Security
“National Security” was mentioned 23 times in the AI Action Plan. This is more than just “data center,” “employment,” “science,” and other important terms. The program’s national security strategy focuses on integrating AI into US defense equipment and intelligence reporting agencies and building DOD’s AI data centers while protesting foreign threats.
Among other things, the plan calls for the DOD and Intelligence communities to regularly assess how AI adoption in the US is compared to rivals like China and adapt accordingly, and to assess the risks posed by both domestic and enemy AI systems.
Within the DOD itself, the strategy emphasizes increasing the skills of military workers, automating workflows, and ensuring priority access to calculate resources in times of national emergency.
Source link
