Nuclear power developers need those rarest of creatures: rich folks prepared to wait. That goes double for the more exotic niche of small modular reactors (SMRs). Now one developer, Kairos Power LLC, appears to have landed a patient whale in the form of Alphabet Inc’s Google.
Kairos, based in Alameda, California, announced late on Monday it had signed an agreement with its near neighbor under which it would develop and build several SMRs, with Google buying the output beginning in 2030 and reaching full capacity of 500 megawatts in 2035. Kairos received a permit in December last year from the US Nuclear Regulatory Commission to develop a pilot project in Tennessee to demonstrate its molten-salt technology, which might offer safety advantages compared with traditional water-cooled reactors. The US Department of Energy is backing the effort with a US$303 million grant.
Something of a pattern is emerging in the effort to revive nuclear power in the US, whereby technology-sector billionaires, such as Bill Gates, Jeff Bezos and Sam Altman, put money into start-ups at the cutting edge. The technology companies are needed to actually buy the eventual output.
Thus far, this has been confined to contracting power for new datacenters focused on artificial intelligence (AI) from existing nuclear power plants. Those deals make sense for big tech, because they offer zero carbon power that is, above all, available relatively quickly to aid the race for AI dominance.
ROOM TO MOVE
However, they are inherently limited — there are only so many existing, unspoken-for reactors to go around — and also carry the seeds of a backlash, since they are effectively repurposing existing nuclear generation, leaving everyone else on the grid to take other, mostly fossil fuel-based, sources.
The most recent case, involving Microsoft Corp and the Three Mile Island site in Pennsylvania, finessed this by using the tech giant’s purchasing power to justify restarting a recently retired reactor. However again, the scope is limited.
ENTER GOOGLE
Google’s offer to take the power from Kairos’ SMRs offers genuine “additionality,” as the phrase goes, by encouraging the construction of brand new reactors and, potentially, seeding a whole new industry in the US. This was noted in the otherwise relatively sparse announcement. Kairos could tout the contract of a giant, credible off-taker for its promised electricity to raise more money to develop the plants.
Meanwhile, Google gets bragging rights about fostering a new zero carbon source of power for its long-term AI efforts — very useful given the likelihood that a large portion of the expected near-term boom in AI-related electricity demand is likely to be gas-fired.
Meanwhile, Google’s upfront cost appears minimal. Both parties are tight-lipped about terms, including pricing. However, there is no mention of Google making an upfront, equity-like investment in Kairos itself, and given the nuclear industry’s long history of delays and busted budgets, it seems inconceivable that Google has not negotiated some sort of exit clause if Kairos’ design does not work out.
Think of this in terms of a contingent liability. Assume Google offered to pay, say, US$150-US$200 per megawatt-hour — higher than even the premiums paid in recent deals with existing plants — and it implies the first payment for full output of 500 megawatts in 2035 would add up to about US$100 million in today’s money.
MIXED PROSPECTS
That is still a big deal for tiny Kairos and its competitors. Shares of Oklo Inc, the SMR start-up backed by Altman, perked up Monday on renewed expectations of a deal with OpenAI. So, too, did NuScale Power Corp, whose stock was crushed last year by the collapse of its own agreement to supply Utah Associated Municipal Power Systems.
That setback is pertinent here, demonstrating that even a signed power purchase agreement does not guarantee success if price estimates start to spiral out of control.
The most salient fact about SMRs is that none have actually been built and commercially operated in the US. That said, it was always a little strange that this leading edge of the vaunted nuclear renaissance was, with NuScale, predicated on the backing of a group of obscure electricity co-operatives in Utah. Google makes for a likelier consort.
Liam Denning is a Bloomberg Opinion columnist covering energy. A former banker, he edited the Wall Street Journal’s “Heard on the Street” column and wrote the Financial Times’ “Lex” column. This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
A return to power for former US president Donald Trump would pose grave risks to Taiwan’s security, autonomy and the broader stability of the Indo-Pacific region. The stakes have never been higher as China aggressively escalates its pressure on Taiwan, deploying economic, military and psychological tactics aimed at subjugating the nation under Beijing’s control. The US has long acted as Taiwan’s foremost security partner, a bulwark against Chinese expansionism in the region. However, a second Trump presidency could upend decades of US commitments, introducing unpredictability that could embolden Beijing and severely compromise Taiwan’s position. While president, Trump’s foreign policy reflected a transactional
There appears to be a growing view among leaders and leading thinkers in Taiwan that their words and actions have no influence over how China approaches cross-Strait relations. According to this logic, China’s actions toward Taiwan are guided by China’s unwavering ambition to assert control over Taiwan. Many also believe Beijing’s approach is influenced by China’s domestic politics. As the thinking goes, former President Tsai Ing-wen (蔡英文) made a good faith effort to demonstrate her moderation on cross-Strait issues throughout her tenure. During her 2016 inaugural address, Tsai sent several constructive signals, including by acknowledging the historical fact of interactions and
Chinese President Xi Jinping (習近平) has prioritized modernizing the Chinese People’s Liberation Army (PLA) to rival the US military, with many experts believing he would not act on Taiwan until the PLA is fully prepared to confront US forces. At the Chinese Communist Party’s 20th Party Congress in 2022, Xi emphasized accelerating this modernization, setting 2027 — the PLA’s centennial — as the new target, replacing the previous 2035 goal. US intelligence agencies said that Xi has directed the PLA to be ready for a potential invasion of Taiwan by 2027, although no decision on launching an attack had been made. Whether
Taiwanese men doing their alternative military service within Taipei’s and Taichung’s Department of Compulsory Military Service independently reported outlandish incidents of brainwashing. Below is a story related to propagandizing as it manifests overseas. One of my Chinese friends and a group of Taiwanese friends set off together from Sydney, Australia, in a tour group to northern Europe. The travel agency arranged for a Chinese tour guide to lead the group. The guide would start off on a propaganda blitz every afternoon at a set time, singing the praises of the Chinese Communist Party (CCP). The Taiwanese members of the tour seemed