US President Donald Trump is expected to announce his much-anticipated AI action plan at a Washington, DC event hosted by Silicon Valley Insider on Wednesday.
The AI Action Plan should provide a roadmap for the Trump administration’s strategy, priorities and concerns on AI.
The plan is effectively an alternative to Biden AI executive order, the previous administration’s AI strategy that focuses on requiring AI companies to submit safety and security reports and limiting racial or other discriminatory biases in the frontier AI model. Trump abolished Biden’s order within days of the inauguration, claiming that the requirement could be troubling for AI companies and could hinder America’s innovation.
In the first six months, the Trump administration has broadly encouraged efforts to accelerate the development and distribution of American AI technologies. Trump helped Openai, Oracle and Softbank announce the multi-billion-dollar Stargate Data Center project, and the president stripped Nvidia from selling AI chips around the world.
At the same time, Trump’s AI Czar David Sacks chose to fight against tech companies over “awakening” AI, claiming Openai, humanity, and Google are infiltrating left-leaning values into AI chatbots and censoring conservative views.
Some people have already pushed back Trump’s AI plan of action as they allegedly put corporate profits first to the public. On Tuesday, a group of over 90 organisations, including labor, environmental justice and consumer protection, issued an open letter called the People’s AI Action Plan. This presents a set of AI policies that argue that it places the interests of American citizens first and counters what Trump is expected to announce.
“At the expense of our freedom and equality, and the happiness of our workers and families, we cannot write AI and economic rules for Big Tech and Big Oil lobbyists,” the group said in a statement to TechCrunch that it acknowledged the energy needs of Silicon Valley’s AI data centers.
TechCrunch Events
San Francisco
|
October 27th-29th, 2025
Trump’s AI plan of action should promote his administration’s agenda more clearly, but remains unknown to exactly. Trump is expected to share details about the plan at the events hosted by Hill and the Valley Forum, and at the “AI Race Winning” Summit, all events, a podcast he co-hosted when he is not a government official or venture capitalist.
Here’s what you know about the AI Action Plan so far:
Acceleration of American AI
According to a report by Time Magazine, Trump’s AI strategy is expected to focus on three pillars: infrastructure, innovation and global impact.
In the case of infrastructure, the Trump administration reportedly plans to overhaul rules to speed up the development of AI data centers. It aims to help AI companies meet the growing energy needs to train and serve AI models. However, the rise of AI data centers is widely expected to cause energy shortages by the end of the decade, unless energy production does not increase rapidly, by sucking up enormous amounts of energy and water from nearby communities.
According to Time, the president’s infrastructure pillars are expected to include plans to modernize America’s electric grid and add new energy sources to power these data centers.
In terms of innovation, Trump is reportedly planning to use AI action plans to revive conversations about the state’s AI law block (though federal proposals on the issue were overwhelmingly unsuccessful last month). This is part of an effort to reduce barriers to innovation for American AI companies, but could ultimately prevent lawmakers from passing meaningful safety and security standards for AI companies.
As for the pillars of global influence, Trump is expected to develop a strategy to advance adoption of American AI models and chips not only in the US but around the world. Federal officials were surprised by the rise of Deepseek, and other Chinese AI labs, such as Qwen and Moonshot AI, have since become a valuable competitor of Openai. Trump wants American technology to be the global standard.
The Trump administration is also expected to sign a series of AI-related executive orders on Wednesday to advance that goal, according to the Washington Post. Some of these orders clear the path for faster datacenter build-outs, while others encourage exports of American technology.
“Awakened” fight AI
One of the executive orders Trump is scheduled to sign a crackdown on the “awakening” AI model on Wednesday, reported by the Wall Street Journal earlier this week. This order requires AI companies with federal contracts, including Openai, Xai, Google, and humanity, to ensure that the AI model has a neutral and unbiased language.
The “awakening” AI crackdown marks the latest Republican attack on the historically left-leaning crowd in Silicon Valley. Over the past few years, Republicans allegedly looked into social media companies and changed their algorithms to censor conservative voices. Meta’s Mark Zuckerberg recently surrendered to these allegations and re-evaluated the moderation of its Facebook and Instagram content to express more voices.
The key question looms around this executive order is who defines whether the AI model is neutral or biased, and how do you decide. Trump has long said he is a defender of free speech, so setting an executive order may provide for what the AI model can do and can’t say. That said, a Florida judge recently ruled that the AI chatbot is not protected by the First Amendment.
In light of this crack, Openai and other AI labs have sought to make AI chatbots represent more perspectives. These companies are in awkward position to generate AI responses that please everyone, but do not spread the perspectives and misinformation of extremists.
Elon Musk, Trump’s biggest financial advocate whose relationship with the president has recently become sour, has explicitly started Xai to develop “conflict” AI chatbots, Grok, and combat chat gupto. However, Xai’s efforts to create chatbots like this have not been very successful. In recent weeks, Xai has been forced to repeatedly apologise when AI chatbots have made anti-Semitic rants and consulted Musk’s personal opinions on the issue of hot buttons.
What Silicon Valley and Big Technology Want
In April, the White House announced it had received more than 10,000 public comments from businesses, local governments and nonprofits regarding Trump’s AI action plan.
Openai, Google, Meta and Amazon have effectively used the opportunity to submit a wish list of friendly AI policies that the Trump administration wants to implement.
Many of America’s leading AI model developers should ask Trump to use AI action plans to ensure that training large language models on copyrighted materials is fair and therefore should be allowed.
Such protections can bring significant benefits to many of these companies. Many companies are involved in aggressive litigation with copyright holders in the music, film, news and book industries. These publishers accuse AI companies of illegally training copyrighted works to create AI models, and could underestimate the media in the process.
Meanwhile, Meta asked Trump to protect his open AI model. This is freely available for download online. By releasing the Llama model openly, Meta was able to undercut Openai and Google’s closed products. However, humanity raised concerns about whether open AI models could leak strong technology to bad actors, including China.
Other interest groups, including nonprofits such as The Future of Life Institute, have used the comment period to urge the Trump administration to increase investment in AI research efforts other than commercial entities. This demand comes when the Trump administration and Doge cut funds for American universities drastically. Many of them have been a great powerhouse for scientific breakthroughs in recent decades.
It appears unlikely that Trump’s AI action plan will feature the same safety and security reporting standards that the Biden administration has attempted to impose. However, polls show that most Americans want AI companies to be bound by safety standards.
While some state lawmakers are pushing for bills that provide a safety and security reporting mandate, if they contradict Trump’s AI plan of action, they could face opposition from the Trump administration and Republican lawmakers.
