Trump's Silicon Valley Advisers Target AI 'Censorship'

President-Elect Trump and Silicon Valley's Influence

Donald Trump, as President-elect, has assembled a team of advisors from the technology sector, including prominent figures such as Elon Musk, Marc Andreessen, and David Sacks. These entrepreneurs are providing counsel on technology-related matters and broader policy issues.

AI Development and the Issue of "Censorship"

Regarding artificial intelligence, this group of technologists generally agrees on the importance of accelerating AI development and its widespread implementation across the United States. However, a recurring concern raised within this circle centers on the potential for AI “censorship” by large technology companies.

A New Front in the Culture War

The counsel offered by Trump’s Silicon Valley advisors may position the responses generated by AI chatbots as a new arena for conservative viewpoints to challenge the ongoing disputes with tech corporations.

Defining AI Censorship

The term “AI censorship” describes instances where technology companies influence the outputs of their AI chatbots to align with specific political ideologies or promote particular perspectives. This is often similar to content moderation, though the latter carries a different implication. Achieving accurate AI responses for breaking news and contentious topics remains a continuous challenge, much like refining social media and search algorithms.

Past Criticisms of Big Tech

For the past ten years, conservatives have consistently voiced criticism of Big Tech companies, alleging that they yield to governmental pressures and suppress conservative voices on their social media platforms and services. Interestingly, some tech leaders have begun to publicly adjust their stances.

For instance, Mark Zuckerberg, CEO of Meta, issued an apology to Congress prior to the 2024 election for complying with the Biden administration’s requests to aggressively moderate content related to COVID-19. He subsequently stated that he had made a “20-year political mistake” by assuming excessive responsibility for issues beyond his company’s control, and pledged to avoid repeating such errors.

AI Chatbots: A Greater Threat to Free Speech?

However, Trump’s technology advisors contend that AI chatbots present an even more significant threat to freedom of expression, and potentially a more potent means of controlling discourse. Rather than simply adjusting a search or feed algorithm to favor a particular outcome—like demoting misinformation about vaccines—tech companies can now deliver a single, definitive answer that excludes opposing viewpoints.

Public Statements from Musk, Andreessen, and Sacks

In recent months, Musk, Andreessen, and Sacks have publicly expressed their concerns regarding AI censorship through podcasts, interviews, and social media platforms. While the specifics of their advice to Trump remain unknown, their publicly stated beliefs offer insights into the discussions taking place in Washington, D.C., and Mar-a-Lago.

Andreessen's Warnings

“My belief, and what I’ve been trying to convey to individuals in Washington, is that if social media censorship was problematic, [AI] possesses the potential to be a thousand times more detrimental,” stated Marc Andreessen, co-founder of a16z, in a recent interview with Joe Rogan. He further elaborated to Bari Weiss, “If one were to construct the ultimate dystopian world, it would be one governed by an AI programmed to deceive.”

Andreessen also revealed to Weiss that he has dedicated approximately half of his time since the election to advising Trump’s team on technology and business matters.

Sacks' Perspective as AI and Crypto Czar

“[Andreessen] elucidated the dystopian trajectory we were on with AI,” commented David Sacks, former COO of PayPal and co-founder of Craft Ventures, in a recent post on X following his appointment as Trump’s AI and crypto advisor. “However, the course has shifted, and we are now on a different path.”

Criticism of "Wokeness" in AI

On the “All In” podcast, which Sacks co-hosts with other prominent venture capitalists, Trump’s new AI advisor has frequently criticized Google and OpenAI for allegedly programming their AI chatbots to adhere to politically correct standards. The hosts of the show characterize this as forcing AI to be “politically correct.”

“A concern with ChatGPT initially was that it was programmed to be ‘woke’ and did not provide accurate answers on numerous subjects. The censorship was embedded within the responses themselves,” Sacks explained during a November 2023 episode of All In.

The Challenge of AI Truthfulness

Despite Sacks’ assertions, even Elon Musk acknowledges that xAI’s chatbot sometimes exhibits a greater degree of political correctness than he prefers. This is likely a consequence of training AI models on the vast and often biased data available on the open internet. Nevertheless, Sacks is increasingly emphasizing the importance of “AI truthfulness.”

The Controversy Surrounding AI Image Generation and Bias

A widely discussed instance of potential AI censorship involved Google’s Gemini, where the image generator produced racially diverse depictions for prompts like “U.S. founding fathers” and “German soldiers in WWII,” despite historical inaccuracies.

However, instances of companies influencing search results aren’t isolated. Recently, it was discovered that ChatGPT declines to respond to inquiries concerning specific individuals, a behavior OpenAI attributes to activated internal privacy protocols.

AI Chatbot Responses to Sensitive Topics

Furthermore, both Google’s and Microsoft’s AI chatbots initially avoided stating the winner of the 2020 U.S. presidential election. During the 2024 election cycle, nearly all AI systems similarly refrained from answering questions about election outcomes, with Perplexity and Grok being notable exceptions.

Tech companies often justify these actions as responsible choices to ensure user safety. In certain cases, this rationale may hold merit—Grok, for example, prematurely predicted the outcome of the 2024 election before any votes were cast.

The Gemini situation, however, proved particularly noteworthy, leading Google to temporarily disable Gemini’s ability to generate images of people—a restriction that remains in place for the free version. Google characterized this as an oversight and issued an apology for the inaccurate results.

Venture Capitalist Perspectives on AI Bias

Marc Andreessen and David Sacks dispute Google’s explanation. Both venture capitalists contend that Google’s actions weren’t accidental, but rather a deliberate demonstration of underlying biases. They view the incident as a revealing moment for the company.

Sacks, during a February 2024 appearance on All In, asserted that the individuals developing Google AI are embedding their own preferences and biases, which lean heavily to the left. He expressed concern that these biases wouldn’t be eliminated, but instead become more subtle.

Andreessen echoed this sentiment in a recent interview with Weiss, reiterating the Gemini incident and stating, “This goes directly to Elon’s argument, which is that at the core of this, you have to train the AI to lie [i.e., to produce answers like Gemini’s].”

Elon Musk and the Pursuit of “TruthGPT”

As Andreessen points out, Elon Musk has been vocal in his criticism of “woke AI chatbots.” In 2023, Musk founded xAI, a well-funded AI startup, as a direct response to OpenAI’s ChatGPT, which he claimed was influenced by a “woke mind virus.” This ultimately led to the creation of Grok, an AI chatbot with fewer restrictions than its competitors.

“I’m going to start something which you call TruthGPT or a maximum truth-seeking AI that tries to understand the nature of the universe,” Musk stated in a 2023 interview with Fox News.

Upon the launch of Grok, Sacks praised the initiative, noting that it would “at a minimum — keep OpenAI honest and keep ChatGPT honest,” as he stated on an All In episode in November 2023.

Musk is now significantly expanding his efforts, having secured over $12 billion in funding for xAI to compete with OpenAI. He is also pursuing legal action against OpenAI and Microsoft, potentially hindering OpenAI’s transition to a for-profit model.

Political Implications and Potential Investigations

Musk’s influence with conservative political figures has already been demonstrated in other contexts. Texas Attorney General Ken Paxton is currently investigating advertisers who allegedly boycotted Elon Musk’s X platform. Musk previously filed a lawsuit against the same advertising group, and some companies have since resumed advertising on his platform.

The specific actions that Donald Trump and other Republicans might take to investigate potential AI censorship by OpenAI or Google remain unclear. Possible approaches include investigations by relevant agencies, legal challenges, or simply making the issue a focal point during the next four years. Regardless, Trump’s advisors in Silicon Valley are being forthright about their concerns.

“Elon, with the Twitter files, did a privatized version of what now needs to happen broadly,” Andreessen told Weiss, referencing Musk’s claims of censorship at Twitter. “We, the American population, need to find out what’s been happening all this time, specifically about this intertwining of government pressure with censorship… There needs to be consequences.”