Reid Hoffman on Trade, AI, Energy and the Incoming Trump Administration

In this December 2024 conversation, investor and entrepreneur Reid Hoffman discusses the incoming Trump administration, trade, AI, energy and venture capital with Harry Stebbings. His policy proposals and predictions are his own views at the time. Hoffman also discloses investments in several technologies and companies discussed.

Hoffman says reports that he was moving to the UK overstated his plans: he remained based in Seattle but wanted to spend more time in Britain. He says he will continue investing in US businesses and hopes to contribute to prosperity even though he does not expect a role in the incoming administration.

He expects to support some goals, such as enabling energy and AI businesses, while criticizing policies he considers harmful. He argues that Americans should be able to criticize those in power, and voices concerns about potential misuse of state authority. These are his political views and concerns, not findings established by the interview.

Asked what could go wrong, Hoffman raises the possibility of politically selective enforcement and financial conflicts involving people with political power. These are hypothetical risks he identifies; the conversation does not establish that any specific transaction or future policy is corrupt.

He argues that broad trade wars could reduce prosperity, while targeted tariffs or sanctions may be appropriate in response to particular conduct by other countries. His examples include reciprocal market access and goods produced with forced labor. These are policy judgments, not a forecast proven by the interview.

On Chinese electric vehicles, Hoffman agrees with Stebbings that subsidies can create a difficult competitive environment and argues for a response if a country seeks to undermine another industry. His characterization of intent is an opinion; the discussion does not establish the motives of every manufacturer.

On possible positive outcomes, he favors simplifying regulation, expanding energy supply and supporting businesses across the economy. He hopes policies for fission, fusion and AI infrastructure can help create future industries.

Hoffman says Elon Musk’s political role could help AI if it expands energy, data centers and talent access across the industry. He also worries that political attention could be diverted to what he calls less central cultural disputes. Both outcomes are conditional on decisions not yet made at the time.

He supports examining government efficiency but doubts that the announced goal of cutting $2 trillion could be achieved merely by reducing federal staffing. He points to large budget areas such as defense and social programs. His judgment is a forecast and the interview does not provide a full budget calculation.

Hoffman argues US defense spending should decrease and says Europe should contribute more to its own defense. He criticizes aircraft-carrier spending and makes claims about missile costs and vulnerability; the interview does not verify those military comparisons.

He supports Ukrainians’ right to self-defense and self-determination and favors supporting them. These are his positions on the war, not an account of all perspectives or a determination of legal responsibility.

On Medicare, he proposes examining the cost and quality of care, including spending near the end of life and prevention. He recognizes that the issue is politically difficult. The comments are policy proposals, not individualized medical guidance.

On TikTok, Hoffman argues for reciprocal commercial treatment if China bars US social-media competitors, while distinguishing this from questions about governance and transparency for platforms globally. Stebbings challenges whether reciprocal bans are a productive response.

Hoffman hopes nuclear regulation can change to enable more fission and eventually fusion. He says he has invested in both areas largely for philanthropic reasons and does not assume a financial return; that involvement is relevant to his advocacy.

He worries that climate change may receive less attention amid AI and global conflict. He hopes investment in clean energy, including nuclear technologies, could counter that trend. These are expectations about policy and technology at the time, not guaranteed outcomes.

Hoffman argues AI could help climate efforts by improving efficiency and encouraging investment in clean power. He gives rough estimates of data-center electricity use and an example of DeepMind improving data-center efficiency. The interview does not establish a complete lifecycle comparison of AI’s energy costs and benefits.

He says large technology customers can help finance new clean-energy supply by committing to buy power. He also describes AI-assisted fusion research as possible but uncertain. Those examples do not prove AI will be a net energy or emissions saver.

Hoffman is enthusiastic about fusion as a possible source of abundant low-carbon energy, including for carbon removal. Commercial fusion and large-scale carbon removal remain future possibilities in his account; the claim that fusion could solve climate change is his technological vision, not an established result.

Discussing Google’s Willow quantum announcement, Hoffman calls it progress but says he does not know the timetable for useful large-scale quantum computing. He offers approximate logical-qubit thresholds for chemistry and encryption; the interview does not substantiate those thresholds or claim current systems meet them.

Stebbings asks whether Musk has unprecedented power. Hoffman says modern institutions need checks, accountability and conflict-of-interest safeguards, such as separating public responsibilities from private business interests. He simultaneously credits Musk with major contributions to electric vehicles and space technology.

Hoffman argues that Musk’s conviction can be valuable when pursuing ambitious projects but can also contribute to amplifying misinformation. He speculates about the origin of some online content; the interview does not establish those particular claims. His assessment combines praise and criticism.

On speech and social platforms, Hoffman distinguishes freedom to express an opinion from how widely it is amplified. He favors a democratically agreed approach that could place expert context beside claims across media, while preserving the ability to express dissent.

Asked about Jeffrey Epstein, Hoffman says he was not a client and describes his contact in connection with MIT fundraising. He says he relied on institutional vetting, regrets that association and has apologized for the reputational benefit it could have given Epstein. The interview presents his account and does not independently resolve broader allegations or supply the underlying records.

Hoffman says he has worked on the intersection of biology and computing and serves on a Chan Zuckerberg Initiative Biohub board. He is optimistic that AI could help drug discovery, including cancer research. This is a research and investment interest and a prediction, not a claim that a specific cancer treatment has been proven effective.

Stebbings asks whether building an AI sales tool is an injustice compared with working on cancer or climate. Hoffman rejects that broad judgment: he wants more people tackling major problems but says improving business processes also contributes to society. He would still celebrate major medical or energy advances as exceptional achievements.

He is optimistic about the UK’s talent, universities and global perspective, while acknowledging economic challenges and criticizing Brexit. He says entrepreneurs need collective investment and support; his support for Entrepreneur First is relevant to that view.

Hoffman favors diversified semiconductor manufacturing, ideally with several sources and at least some capacity in Western countries. He supports the US CHIPS Act as a way to rebuild capacity and views dependence on a single location as a resilience risk. This is a policy preference, not an assessment of the economics of a particular facility.

He expects Nvidia to remain strong while competitors also build chips, citing high demand for compute. His assessment is a market forecast, not a guarantee of Nvidia’s future share or returns.

Hoffman disputes the claim that large-language-model scaling has reached its limit. He expects larger training runs, data and other improvements to continue producing new capabilities, while also expecting many agents built from multiple models. These are his predictions about a fast-moving field.

He says a technological boom can contain both poor investments and transformative companies. Investors must judge probabilities rather than assume every expensive deal is foolish or every new category succeeds.

He recalls leading Airbnb’s Series A at a $60 million post-money valuation when transaction volume was still small. His investment thesis depended on the possibility of transforming an industry, and he says future financing needs still mattered to evaluating price. He had a financial interest in Airbnb.

On acquisitions, Hoffman argues that blanket restrictions on large-company acquisitions could discourage investors from funding challengers because an exit path may disappear. He acknowledges that some specific acquisitions can merit scrutiny. This is his antitrust theory, not a legal conclusion.

He discloses that he was a Figma investor and says he may make more money because the proposed transaction did not proceed, while criticizing the regulatory theory behind it. That financial interest matters when assessing his position.

In quick fire, he predicts AI will create value for both large and small companies, and says he considers Twitter worse after Musk’s ownership. He would pursue fusion energy if he knew he could not fail. These are opinions and aspirations.

He would like venture investment to cover more physical technologies as well as software, citing energy and transportation. He says he owns Bitcoin and has invested in cryptocurrency companies; his discussion of how governments should engage with crypto reflects that interest.

Hoffman describes recent congressional testimony about unidentified aerial phenomena as an open question that prompted curiosity. He explicitly does not claim the testimony establishes extraterrestrial technology and says there may be other explanations or nothing to the claims.

He declines to give a precise Bitcoin price forecast, says his investment horizon is seven to ten years and discloses that he holds Bitcoin. He describes missed follow-on investment in Airbnb and missed early opportunities in Stripe and Square as financial regrets. These are retrospective personal judgments, not investment advice.

The advice from his father that stays with him is that making a decision narrows immediate options but creates the possibility of long-term opportunity. He closes by emphasizing the need to choose rather than indefinitely preserve every option.