Trump's AI Strategy: Prioritizing Growth Over Regulation

A New Direction for AI: The Trump Administration's Action Plan

On Wednesday, the Trump administration released its long-awaited AI Action Plan. This document represents a significant departure from the more measured strategy previously adopted by former President Biden regarding the potential risks associated with artificial intelligence.

Instead of prioritizing caution, the plan aggressively pursues the expansion of AI infrastructure, the reduction of regulatory burdens for technology firms, the strengthening of national security measures, and enhanced competition with China.

Potential Impacts Across Sectors

The consequences of this strategic shift are anticipated to be widespread, impacting numerous industries and potentially influencing everyday consumers. A key aspect of the AI Action Plan is its diminished focus on mitigating the potential negative effects of AI.

The plan instead emphasizes the construction of data centers to support the AI industry, even if this necessitates utilizing federal lands or maintaining power supply during critical times for the energy grid.

Implementation and Future Steps

The ultimate impact of the AI Action Plan will largely depend on its execution, with many specific details still under development. The document functions more as a strategic outline than a detailed set of instructions.

However, the overarching objective is clear: prioritizing advancement and innovation above all else. The administration believes this approach is essential for progress.

Justification and Key Goals

The Trump administration frames this initiative as the only path to “usher in a new golden age of human flourishing.” A central goal is to persuade the public that investing billions of dollars in taxpayer funds for data center construction is beneficial.

The plan also incorporates proposals for workforce development, including initiatives to upskill workers and collaborate with local governments to generate employment opportunities within the data center sector.

“To secure our future, we must harness the full power of American innovation,” stated Trump. “We will continue to reject radical climate dogma and bureaucratic red tape. Simply put, we need to ‘Build, Baby, Build!’”

Authorship and Public Input

The AI Action Plan was developed by the Trump administration’s team of technology and AI experts, many of whom originate from Silicon Valley companies.

Key contributors include Michael Kratsios, director of the Office of Science and Technology Policy; David Sacks, designated as the AI and crypto czar; and Marco Rubio, assistant to the president for national security affairs.

The plan’s formulation was informed by over 10,000 public comments submitted by various interest groups.

The Shifting Landscape of AI Regulation and the Proposed Moratorium

Early this month, the Senate rescinded a debated clause within the budget legislation that aimed to prevent states from enacting AI regulations for a decade. Had this provision remained, states’ access to federal broadband funding would have been contingent upon adherence to the proposed moratorium.

However, the issue remains unresolved, as the AI Action Plan investigates alternative methods to restrict state-level AI regulation. Driven by a wider objective to “stimulate economic growth through deregulation,” the administration is now considering limiting federal funding to states based on the AI laws they implement.

The plan further instructs the Federal Communications Commission to assess whether state AI regulations impede the agency’s capacity to fulfill its duties and responsibilities. Essentially, if state AI rules impact radio, television, or internet services – a common occurrence – the FCC could potentially intervene.

At the national level, the action plan tasks the Office of Science and Technology Policy with soliciting feedback from businesses and the public regarding existing federal regulations that may obstruct AI innovation and implementation. This input will then inform subsequent actions by federal agencies.

Key implications of this plan involve a potential shift in power, moving regulatory control of AI away from individual states and towards the federal government. This could lead to a more uniform, but potentially less responsive, regulatory environment.

Further Details of the AI Action Plan

The administration’s approach centers on the belief that excessive regulation stifles AI innovation. By reducing regulatory burdens, the goal is to accelerate the development and adoption of AI technologies.

The plan specifically targets regulations that could create barriers to entry for smaller companies or limit the scope of AI applications. This includes examining rules related to data privacy, algorithmic transparency, and liability for AI-driven decisions.

• The FCC’s role is crucial, as it oversees the communication infrastructure that underpins many AI applications.
• The Office of Science and Technology Policy will serve as a central coordinating body for gathering information and formulating policy recommendations.
• The potential for reduced federal funding creates a significant incentive for states to reconsider their AI regulatory approaches.

Streamlining Regulations for Data Centers

The current administration, under Trump’s direction, is pursuing deregulation to expedite the development of infrastructure crucial for Artificial Intelligence. This includes data centers, semiconductor fabrication facilities, and supporting power generation. The administration contends that current environmental regulations – notably NEPA, the Clean Air Act, and the Clean Water Act – are impeding the nation’s ability to keep pace with the escalating demands of AI development.

Consequently, Trump’s AI Action Plan prioritizes the stabilization of the American energy grid. Simultaneously, the plan directs the federal government to explore innovative methods for managing power usage by substantial consumers, specifically AI-driven companies, during times of peak grid stress.

Several organizations, including xAI and Meta, have faced scrutiny regarding the concentration of pollution within susceptible communities. Accusations have been leveled against xAI, alleging circumvention of environmental protections and the exposure of local populations to detrimental emissions from gas turbine generators at their Memphis data center.

The proposed action plan advocates for the implementation of categorical exclusions, the simplification of permitting procedures, and the broadened application of expedited programs such as FAST-41. These measures aim to facilitate the construction of vital AI infrastructure, particularly on federally owned lands.

This includes national parks, designated wilderness areas, and military installations. Reflecting a recurring theme in Trump’s policies – competition with China – the strategy emphasizes restricting foreign technology and bolstering security measures.

Security protections are intended to prevent the introduction of “adversarial technology,” such as chips and hardware manufactured in China, into the U.S. supply chain.

Key Components of the Plan

• Deregulation: Reducing the burden of environmental regulations on AI infrastructure projects.
• Grid Stabilization: Strengthening the national energy grid to support increased power demands.
• Permitting Reform: Streamlining the approval process for building data centers and fabs.
• Supply Chain Security: Protecting the U.S. technology supply chain from foreign adversaries.

The plan seeks to accelerate the deployment of AI infrastructure by addressing perceived bottlenecks in the regulatory process. This is viewed as essential for maintaining a competitive edge in the global AI landscape.

Furthermore, the emphasis on supply chain security underscores concerns about the potential for malicious actors to exploit vulnerabilities in the technology supply chain. The administration aims to mitigate these risks through stricter controls and domestic production incentives.

Trump’s Stance Against “Biased AI”

A central component of Trump’s AI Action plan centers on safeguarding free speech and upholding “American values.” This is partially achieved through the removal of references to misinformation, Diversity, Equity, and Inclusion (DEI) initiatives, and climate change from federal risk-assessment frameworks.

The plan explicitly states the necessity for these systems to be developed with freedom of speech and expression as foundational principles. It further emphasizes that U.S. government policy should not impede this objective.

The intention is to guarantee that government policy does not infringe upon free speech; however, the AI Action Plan itself carries the potential to do so.

A key policy recommendation involves revising federal procurement guidelines. This would ensure the government only contracts with developers of leading large language models who can demonstrate their systems are objective and devoid of ideological bias imposed from above.

This phrasing aligns with reports in The Wall Street Journal regarding the content of Trump’s forthcoming executive order, anticipated to be released soon.

A significant challenge lies in the difficulty of achieving true objectivity. Currently, the government has not established a clear methodology for evaluating models based on neutrality.

“Complete neutrality would essentially require a complete lack of interaction,” explains Rumman Chowdhury, a data scientist, CEO of Humane Intelligence, and former U.S. science envoy for AI, in a statement to TechCrunch.

Companies including Anthropic, xAI, Google, and OpenAI have already secured government contracts valued at up to $200 million each. These contracts are intended to facilitate the integration of AI applications within the Department of Defense. The potential consequences of Trump’s policy proposals and the upcoming executive order could be substantial.

“For example, a directive prohibiting any business with companies producing AI models deemed ‘non-neutral’ would likely be considered a violation of the First Amendment,” notes Eugene Volokh, a legal scholar specializing in First and Second Amendment law, in an email.

Volokh continues, “An order stipulating that contracts will only be awarded to models demonstrating sufficient neutrality would be more legally sound, although effective implementation remains challenging. This is largely due to the inherent difficulty in defining ‘neutrality’ in these contexts.”

He further suggests, “If the order directs agencies to prioritize both accuracy and neutrality when selecting AIs, granting each agency some discretion in interpreting these criteria, it may prove more legally viable.”

Fostering an Open Ecosystem for Artificial Intelligence

The AI Action Plan proposed by Trump intends to promote the creation and implementation of openly accessible AI models. These models, available for free download, are designed to reflect American principles and values.

This initiative appears to be largely motivated by the increasing prominence of open AI models originating from Chinese AI research institutions, such as DeepSeek and Alibaba’s Qwen.

A key component of the plan involves guaranteeing that startups and researchers engaged in developing open models have access to substantial computing resources.

Such resources are typically costly and historically limited to large technology corporations capable of securing significant contracts—worth millions or even billions of dollars—with cloud service providers.

Furthermore, Trump has stated his intention to collaborate with prominent AI model developers to broaden the research community’s access to both proprietary AI models and relevant data.

American companies and organizations already committed to an open-source approach, including Meta, AI2, and Hugging Face, stand to gain from Trump’s support for open AI development.

Benefits of Open AI

• Increased accessibility for researchers and startups.
• Alignment with American values in AI development.
• Potential to counter the influence of foreign AI models.

Open AI models empower a wider range of innovators to contribute to the field, fostering a more diverse and competitive landscape.

By prioritizing access to computing power and data, the plan aims to level the playing field and accelerate progress in artificial intelligence.

AI Safety and Security Considerations

The AI Action Plan proposed by Trump incorporates measures designed to address concerns within the AI safety community. A key component of this plan is the initiation of a federally funded research and development initiative. This program will concentrate on areas such as enhancing AI interpretability, developing robust AI control systems, and improving adversarial robustness.

Furthermore, Trump’s plan directs various federal agencies, notably the Department of Defense and the Department of Energy, to organize hackathons. These events will be specifically geared towards identifying potential security weaknesses within their respective AI systems.

The plan also recognizes the potential for AI systems to be exploited in malicious activities. This includes their possible contribution to the escalation of cyberattacks, as well as the creation of advanced chemical and biological weapons.

Consequently, the plan requests that developers of cutting-edge AI models collaborate with federal agencies. The purpose of this collaboration is to assess these risks and determine how they might compromise the national security of the United States.

In contrast to Biden’s AI executive order, Trump’s plan demonstrates a reduced emphasis on mandating comprehensive safety and security reporting from leading AI model developers. Many technology companies have expressed that such reporting requirements are unduly burdensome, a sentiment that appears to be acknowledged by Trump’s approach.

Key Differences in Approach

• Focus on Research: Trump prioritizes research into AI safety mechanisms.
• Hackathons for Security: Proactive vulnerability testing through hackathons is emphasized.
• Risk Evaluation Collaboration: A call for cooperation with developers to assess national security threats.
• Reduced Reporting Requirements: A lighter regulatory touch regarding mandatory safety reporting.

The differing strategies reflect contrasting philosophies regarding the regulation and oversight of rapidly evolving AI technologies.

Restricting Access to China

Consistent with expectations, Trump is integrating his existing trade conflict with China into the development of artificial intelligence, as evidenced by his proposed action plan.

A significant component of Trump’s AI Action Plan centers on restricting access to sophisticated AI technologies for entities perceived as posing “national security” risks.

The plan stipulates that various federal agencies will collaborate to gather intelligence regarding overseas, cutting-edge AI initiatives that might jeopardize U.S. national security interests.

Specifically, the Department of Commerce has been assigned the responsibility of analyzing Chinese AI models to determine their consistency with the narratives promoted by the Chinese Communist Party and the extent of censorship present within them.

Furthermore, these entities will undertake evaluations to ascertain the degree to which AI is being implemented by nations considered adversaries of the United States.

These assessments will provide insights into potential vulnerabilities and strategic advantages held by opposing forces.

Key Elements of the Plan

• Intelligence gathering on foreign AI projects.
• Evaluation of Chinese AI models for alignment with CCP messaging.
• Assessment of AI adoption rates among adversaries.

The overarching goal is to safeguard American technological superiority and mitigate potential threats arising from the proliferation of advanced AI capabilities.

National Security and Artificial Intelligence

The phrase “national security” appears prominently within the AI Action Plan, being referenced a total of 23 times.

This frequency surpasses mentions of critical elements such as “data centers,” “jobs,” and “science,” highlighting its importance.

The core of the plan’s national security approach involves the integration of AI into the United States’ defense and intelligence infrastructure.

Furthermore, it proposes the development of dedicated AI data centers specifically for the Department of Defense (DoD), alongside measures to mitigate potential threats originating from abroad.

Strategic Assessment and Adaptation

The plan mandates regular evaluations by the DoD and the intelligence community.

These assessments will compare the rate of AI adoption within the U.S. to that of competitor nations, notably China, and will inform necessary adjustments to strategy.

A key component of this strategy is the ongoing evaluation of risks presented by both domestically developed and adversary AI systems.

Priorities Within the Department of Defense

The DoD’s internal strategy places significant emphasis on enhancing the skills of its personnel through AI-focused training programs.

Automation of existing workflows is also a priority, aiming to improve efficiency and effectiveness.

To ensure operational readiness, the strategy also advocates for prioritized access to computing resources for the DoD during times of national crisis.

Confidential information or sensitive insights regarding the AI industry are welcome. We are dedicated to reporting on the internal operations of this evolving field, from the companies driving innovation to the individuals affected by their choices. Please contact Rebecca Bellan at rebecca.bellan@techcrunch.com and Maxwell Zeff at maxwell.zeff@techcrunch.com. Secure communication is available via Signal at @rebeccabellan.491 and @mzeff.88.