Over Labor Day weekend, social media lit up with observations that President Trump had not been seen in public for several days. Soon, rumors swirled about Trump’s health—and ghoulish hashtags even claimed he had died. Yes, it was just another weekend in the online rumor mill, but this round of speculation came with a novel twist: a flurry of bets about the president’s health on so-called prediction market sites. On Kalshi, the odds of Vice President JD Vance taking office by the end of the year shot up to 15%. For Kalshi customers, a wager of $15 would mean a payout of $100 if Vance took office.
Trump’s alleged disappearance, of course, proved a false alarm. By Tuesday, the internet had moved on to other diversions—but not before pundits blasted Kalshi and its prediction markets rival Polymarket for running “assassination markets,” where the public could (indirectly) wager on the death of a public figure.
Those accusations may have been overblown—not least because one of Trump’s sons invests in and advises both Kalshi and Polymarket. But the episode showed how prediction markets, long the province of a niche clique of academics, have suddenly become a mainstay of politics and the news cycle.
They are also on the cusp of becoming big business.
Kalshi and Polymarket have been around for seven and five years respectively, but their big breakout came during last year’s U.S. presidential election campaign. Over the course of several months, millions of people convened on the platforms to wager more than $3 billion on the outcome, resulting in forecast that proved far more accurate than the most highly regarded polls. For the startups’ founders, this proved their thesis: that the platforms’ blend of crowdsourced wisdom and financial self-interest offers an unprecedented window into future events.
Right now on Kalshi and Polymarket, those future events include profound geopolitical and economic questions, like whether China will invade Taiwan by the end of 2025 (6% as of mid-September) or how many rate cuts the Fed will implement by end of year (14% for two cuts). There are also plenty of more frivolous wagers, like whether Taylor Swift will get pregnant in 2025 (15%).
To their backers, these wagers (“events contracts” in prediction markets parlance) represent a promising new industry—and a potentially powerful tool that investors could use to hedge their portfolios, or that businesses could use to predict consumer demand. Sequoia venture capitalist Alfred Lin describes the markets to Fortune as “basically truth machines.”
Seizing on a favorable environment for fintech experimentation, Lin and others have poured hundreds of millions of dollars into Kalshi and Polymarket (each of which now enjoy $1-billion-plus “unicorn” valuations) and a handful of smaller platforms; public companies like Robinhood are also jostling for a piece of the action. Right now, monthly wagers on Polymarket and Kalshi are totaling well over $1 billion, while analytics firm Similarweb says the sites attracted over 35 million visitors this summer.
Still, the emerging sector is fraught with risks. While their supporters envision prediction markets as nimble tools for peering into the future, many others—including, it seems, most of the people actually using them—see them as just another way to gamble. If the markets come to be seen primarily as just another casino, they are likely to lose the moral and intellectual high ground their boosters have touted. That’s not to mention the challenge of the brutal competition and legal jeopardy that would go with operating in the tightly regulated gaming industry.
There’s also the public unease around platforms that permit wagers on war or the health of politicians in a time of general social upheaval—an unease intensified by the murder of conservative activist Charlie Kirk. And some question whether the two leading prediction market companies can be trusted to run their startups responsibly. In the past year, the founders of both Kalshi and Polymarket have engaged in eyebrow-raising antics that could give pause to investors and regulators alike.
The biggest risk hanging over the industry, though, is a basic business question: Can sites like Kalshi and Polymarket generate sustained interest—and revenue—outside of the once-in-four-years presidential contest?
Prior to election night, nearly $211 million of prediction market bets came flooding in. The following day, gleeful winners lined up to collect—pleased to have anticipated the triumph of the Democratic incumbent, Woodrow Wilson.
That was in 1916, a year that would prove to be the high-water mark of U.S. prediction markets for over a century. When these markets surged back to prominence in last year’s Trump-Harris contest, they were clothed in Kalshi and Polymarket’s digital wrapper. But the underlying mechanics are very much the same as in Wilson’s era.
You can think of prediction markets as wagers that are fluid. Unlike casinos or conventional betting sites, where bettors place a fixed wager against the house, participants on Kalshi and Polymarket bet against one another and can close out their “event contract” anytime. Like stock exchanges, prediction markets serve as a matching service between buyer and seller.
For instance, a contract for a heavily favored election candidate might cost 80¢, which locks in a $1 payout if the candidate wins. The opposing bettor buys a 20¢ ticket that pays $1 if the other candidate wins. But if the favored candidate suffers a major scandal, the value of that contract might drop to 40¢—leaving the owner to decide whether to sell it to another bettor or hold it till the election results come in.
While this is a form of gambling, proponents argue that any negative social effects are outweighed by the powerful signals the contracts can provide to markets about everything from weather in harvest season to whether a given politician will be elected. In practice, the closer to $1 the value of the event contract comes, the greater the likelihood of the event coming to pass. As the Trump election results demonstrated, the markets can be uncannily accurate—and the more people participate, the more accurate they theoretically become.
According to Kalshi cofounder Tarek Mansour, that accuracy is the result of two interlocking factors. “They’re a market-based mechanism, so you get the wisdom of the crowds,” he explains. “Number two, skin in the game. When people have real money on the line, they don’t lie.
Why did something so useful fall out of favor in the first place? The best answer is that prediction markets got swept up in broader Progressive Era campaigns against gambling, just as the rise of scientific polling pioneered by George Gallup provided a useful alternative. (Ironically, the anti-gambling crusades of that era often spared horse racing, since, in the eyes of the moralists, it enjoyed an association with rural American virtue—while also teaching young, military-age men to size up horseflesh.)
Robin Hanson, a George Mason University professor, sees the debate over prediction markets as part of a longtime push and pull between those who view tools for speculation as a moral threat, and those who see them as useful. “Moralizing about betting markets goes up and down in cycles,” Hanson observes. “Pretty much all financial markets were illegal at some point, including stocks and life insurance.”
In this view, prediction markets are taking their place next to products like options and futures contracts, which regulators long frowned on as overly speculative, but are now viewed as important market signals.
Lin of Sequoia is a Kalshi board member who studied prediction markets in college, and he believes they offer a superior way to hedge against uncertainty, allowing investors to fortify themselves against, for example, adverse interest rate movements. “Right now, the way to do that is to look for interest-sensitive stocks and either buy or short them,” says Lin. “There needs to be a better way.”
Mansour says first-hand experience led him to the same conclusion; he once worked on the “exotics” desk of Goldman Sachs, where he built baskets of stocks to help customers take positions on events like Brexit.
This push to open prediction markets follows decades of the U.S. banning them—though not entirely. In 1998, the Commodity Futures Trading Commission (CFTC) permitted an entity called the Iowa Electronic Markets to run betting platforms in which a small group of academics could wager very small amounts. The agency then gave permission in 2014 to a successor group, PredictIt, to operate a somewhat broader version.
In the past two years, the legal chains holding back prediction markets have largely vanished. But the current era of these markets is being shaped by startup founders with a penchant for bending the rules.
“Bruh!” Shayne Coplan’s curly head yells on my iPhone screen shortly before last year’s election. It’s clear Coplan is sore over a Fortune story that revealed that many of the wagers on Polymarket had come from fishy trades.
That revelation didn’t dim the general enthusiasm around Polymarket, and in the subsequent months, I repeatedly propose Zoom or in-person meetings so Coplan can tell me the full story. Nope: Coplan prefers to do it his way, with direct messages and ambush video calls over Signal, where he has chosen the Beatles’ Revolver cover as his avatar.
The 27-year-old Coplan, a New York City kid who got deep into cryptocurrency in high school, started Polymarket in 2020 after dropping out of NYU. One of his backers, Rob Hadick of Dragonfly Capital, describes him as brilliant, with a deep, all-consuming passion for probabilities. Recent accounts and photos of Coplan reflect a founder with a cooler-than-thou affect possessed of deep confidence—or perhaps overconfidence. Recalling encounters with him, two crypto executives told Fortune of Coplan comparing himself to Apple founder Steve Jobs.
Coplan has also brought his swagger to the way he operates his company. In 2022, Polymarket was hit with a CFTC consent decree barring it from operating in the U.S. Despite this, there’s ample evidence that the site turned a blind eye to Americans who placed bets by using a VPN to mask their location. (Polymarket disputes this characterization.) It’s quite possible that this conduct—or Polymarket’s decision to pay U.S. influencers to promote the site—explained the Justice Department’s decision shortly after the election to raid Coplan’s apartment and seize his smartphone and other devices.
The sensible response to a federal raid is to let your lawyers do the talking. Coplan chose another strategy. Days later, he took to X to tweet “New phone who dis?” The gesture amounted to taunting the prosecutors but would ultimately do him no harm; months later, the Feds dropped the investigation without filing charges.
If Coplan relishes being the enfant terrible of prediction markets, Kalshi’s cofounders have taken a different tack. Mansour and cofounder Luana Lopes Lara are eager to talk up their track record of compliance. This included staying well clear of the U.S. market until September 2024, when Kalshi prevailed in a lawsuit against the CFTC, with a federal judge ruling the agency lacked jurisdiction over events contracts unless they concerned “assassination,” “terrorism,” or “gaming.”
Mansour, 29, is slightly disheveled and shares little in common with Coplan save for a fixation with probabilities. Born in California, the Kalshi CEO returned with his parents to a Christian village in Lebanon as a young child. “We went through a few periods of war or terrorism. It was an anxiety-inducing period,” recalls Mansour, adding that he responded to the turmoil by becoming obsessed with math, and then with getting into MIT. Today, he posts his 5.0 GPA from the university on his LinkedIn page.
His cofounder has a different story and mien. Lopes Lara, also 29 and an MIT grad, was born and raised in Brazil and became a professional ballerina before abruptly pivoting to mathematics. Polished and easy in conversation, she recalled moments when prediction markets directly intersected with her own life.
The question “will Kalshi or Polymarket win the most market share?” would be great fodder for a crowdsourced answer.
“The Kalshi markets started predicting that [COVID] was going to pick up again around Thanksgiving a couple years ago, and we made our own return-to-office decisions thinking about this,” she recalls. The prediction proved correct, she adds: “It was very cool to see and follow it, since you could predict what was going to happen in the news a week later.” Lopes Lara also remembers taking a keen interest in a more trivial wager over whether the band One Direction would reunite, and realizing to her deep disappointment that they wouldn’t.
The Kalshi cofounders run a tight ship, with Mansour serving as the public face of the company while Lopes Lara runs internal operations. The team closely vets new events contracts and has added rules to address unanticipated or controversial outcomes. Those include the “will Trump leave office” contract: Under Kalshi rules, that contract will pay out only partially in the event the president dies, rather than paying out “yes” bettors in full.
Polymarket has no such provision for its “will Trump leave office” contract. It has also listed other bets that ended in controversy—among them, a recent wager over whether the president of Ukraine would wear a suit at a White House visit. When Volodymyr Zelensky turned up in black raiment that media outlets described as a suit, Polymarket nonetheless chose to pay those who bought “no” contracts. That decision followed a shadowy dispute-resolution process involving a vote among holders of an obscure cryptocurrency—hardly the kind of adjudication that mollifies critics or customers.
Coplan’s site has given rise to other controversies, including its decision in January to list contracts on when the catastrophic fires around Los Angeles would be contained—wagers that detractors blasted as “arson markets.” The site, which relies on crypto-based contracts, has also drawn flak for being a locus of “wash trading,” identified by blockchain forensics firms, which involves transactions in which one person takes both sides. While wash trading is common on many crypto sites as a way for traders to artificially bump up a coin’s liquidity or feign momentum, its presence makes it hard to ascertain the true volume of wagering on Polymarket. (“Polymarket’s Terms of Use expressly prohibit market manipulation,” a company spokesperson said in response to an earlier Fortune article that examined the issue.)
Kalshi has had fewer legal and ethical stumbles than Polymarket, but the startup hasn’t always modeled good corporate behavior. Most notably: The site responded to news of the FBI’s raid on Coplan’s house with a dirty-tricks campaign that paid at least four influencers to post social media comments highlighting the episode. Among other moves, Kalshi asked the former NFL star Antonio Brown to tweet news of the incident along with the comment “this nigga seems guilty,” which Brown promptly did. Soon after, tech news site Pirate Wires published direct messages linking Kalshi employees to the campaign.
At the time, Kalshi declined to condemn the behavior or discipline the employee responsible. When asked about the episode in a recent interview, Lopes Lara expressed regret, saying the person responsible had not informed her or Mansour about the plan. “Everyone makes mistakes,” she said. “That was a mistake; it was over the line. It’s not something we identify with or would do again.”
While Polymarket and Kalshi have skirmished and pushed boundaries, investors have only grown more enthusiastic. This June, Polymarket finalized a $200 million investment led by Peter Thiel’s Founders Fund, valuing the startup at $1 billion. Meanwhile, Kalshi the same month raised $185 million from Sequoia, Paradigm, and others, at a $2 billion valuation. In September, an anonymously sourced report on tech website The Information claimed both companies were raising more money at significantly higher valuations.
That investor enthusiasm coincides with buzz among news outlets and on social media. But that doesn’t mean the platforms are a sure bet as a business.
Polymarket and Kalshi both peaked on Election Day, when Mansour recalls his site eclipsing even Pornhub (the internet’s most popular destination most days). Since then, no wager listed on either site has come close to re-creating the billions of dollars of bets generated by the Trump-Harris contest. Daily app downloads last October topped 100,000 for Kalshi and 50,000 for Polymarket, according to the companies; the respective figures this June were closer to 6,500 and 650, according to Apptopia.
For now, it’s hard to do a head-to-head comparison between the two companies. Kalshi leads in app download figures. Web traffic tells a different story: Similarweb says Polymarket received 31.7 million visitors between June and August while Kalshi received 4.5 million. But Polymarket’s wash-trading phenomenon is very likely inflating its traffic volumes, while Kalshi’s app advantage can be discounted by the fact it has been the only one of the two allowed to operate in the U.S.
For Kalshi, victory in last year’s CFTC case has served as a regulatory moat to give it a competitive edge. For months, the company also appeared to have an additional political ace up its sleeve in Washington, D.C., in the form of Donald Trump Jr., who became a paid advisor in January. Kalshi’s advantages have quickly eroded, however. Polymarket recently acquired a company that will soon enable it to operate in the U.S. without violating its ongoing consent decree. And Don Jr. revealed in August that his venture capital firm, 1789 Capital, had invested in Polymarket, and that he has joined that startup’s advisory board as well.
Prediction markets may also not be a two-horse race for much longer. New competitors include startups like Railbird and one called The Clearing Company—started by former Polymarket executives. Trading giant Robinhood has also jumped into the sector, offering a series of wagers on high-profile sporting events via third-party partners, including but not limited to Kalshi.
There’s also uncertainty around how these firms plan to make money. The most obvious model is by charging commissions: Kalshi charges around 1% on bets by customers, who are currently wagering an average of $19 million per day. For now, Polymarket is charging no fees, though Hadick, the venture investor, says the site could easily earn several hundred million dollars a year if it did so. Both sites are also signing partnerships with media and AI companies that could yield revenue in the form of data licensing or research fees. Crypto could be another revenue stream: Polymarket is rumored to be launching a digital token, and Kalshi is rushing to embrace blockchain.
All of this, though, will depend on the companies creating a critical mass of liquidity for prediction markets—and a growing reputation for accurate predictions—by persuading everyday people to use them. According to Lin of Sequoia, these tools will follow the same trajectory as any other new technology, spreading from early adopters to the broader public as they become more familiar.
Kalshi’s most popular bets since last year’s election night offer a glimpse of how that adoption might occur. Recent buzzy contracts include wagers on the New York City mayoral race and two sporting events. But Kalshi bettors also rushed to place wagers on Trump’s Liberation Day tariffs, and on the resignation of the notorious “kiss cam” Astronomer CEO.
For now, it appears both sites are devoting most of their promotional efforts to sports-related wagers. While that opportunity appears to be low-hanging fruit—Kalshi’s biggest non-election success came in the form of more than $500 million in bets on March Madness—it also may be short-lived. Ordinarily, the sports wagers would be a clear violation of state regulations on gaming, which govern casinos and betting sites like DraftKings. Kalshi and Polymarket are relying on somewhat convoluted legal reasoning that “events contracts” are different from gambling. While Kalshi prevailed in one court ruling, it is not hard to imagine another court finding otherwise—and several lawsuits are proceeding through state courts.
Ultimately, questions like “Will Kalshi or Polymarket win the most market share?” or “Will an appeals court ban sports betting on prediction markets?” would be great fodder for a crowdsourced answer from a large group of people with skin in the game. For now, at least, those are two bets you won’t find on Kalshi or Polymarket.
The bets drawing wagerers to prediction markets
Politics made Kalshi and Polymarket famous, but other topics are attracting big money
$130.5 million 2025 NBA Finals: Oklahoma City or Indiana?
$88.5 million Sept. 2025 Fed rate decision: How big a cut?
$504.2 million March Madness, 2025: Picking NCAA hoops winners.
$218.8 million*
Daily temperatures in multiple cities
$67.8 million* Rotten Tomatoes scores: Ratings on the review site. Recent bets on Kalshi. *Money wagered cumulatively in long-running betting series (as of 9/17/25)
This article appears in the October/November issue of Fortune with the headline “Wanna bet? Why investors are gambling on Kalshi and Polymarket.”
The Committee for a Responsible Federal Budget (CRFB) is a nonpartisan watchdog that regularly estimates how much the U.S. Congress is adding to the $38 trillion national debt.
With enhanced Affordable Care Act (ACA) subsidies due to expire within days, some Senate Democrats are scrambling to protect millions of Americans from getting the unpleasant holiday gift of spiking health insurance premiums. The CRFB says there’s just one problem with the plan: It’s not funded.
“With the national debt as large as the economy and interest payments costing $1 trillion annually, it is absurd to suggest adding hundreds of billions more to the debt,” CRFB President Maya MacGuineas wrote in a statement on Friday afternoon.
The proposal, backed by members of the Senate Democratic caucus, would fully extend the enhanced ACA subsidies for three years, from 2026 through 2028, with no additional income limits on who can qualify. Those subsidies, originally boosted during the pandemic and later renewed, were designed to lower premiums and prevent coverage losses for middle‑ and lower‑income households purchasing insurance on the ACA exchanges.
CRFB estimated that even this three‑year extension alone would add roughly $300 billion to federal deficits over the next decade, largely because the federal government would continue to shoulder a larger share of premium costs while enrollment and subsidy amounts remain elevated. If Congress ultimately moves to make the enhanced subsidies permanent—as many advocates have urged—the total cost could swell to nearly $550 billion in additional borrowing over the next decade.
Reversing recent guardrails
MacGuineas called the Senate bill “far worse than even a debt-financed extension” as it would roll back several “program integrity” measures that were enacted as part of a 2025 reconciliation law and were intended to tighten oversight of ACA subsidies. On top of that, it would be funded by borrowing even more. “This is a bad idea made worse,” MacGuineas added.
The watchdog group’s central critique is that the new Senate plan does not attempt to offset its costs through spending cuts or new revenue and, in their view, goes beyond a simple extension by expanding the underlying subsidy structure.
The legislation would permanently repeal restrictions that eliminated subsidies for certain groups enrolling during special enrollment periods and would scrap rules requiring full repayment of excess advance subsidies and stricter verification of eligibility and tax reconciliation. The bill would also nullify portions of a 2025 federal regulation that loosened limits on the actuarial value of exchange plans and altered how subsidies are calculated, effectively reshaping how generous plans can be and how federal support is determined. CRFB warned these reversals would increase costs further while weakening safeguards designed to reduce misuse and error in the subsidy system.
MacGuineas said that any subsidy extension should be paired with broader reforms to curb health spending and reduce overall borrowing. In her view, lawmakers are missing a chance to redesign ACA support in a way that lowers premiums while also improving the long‑term budget outlook.
The debate over ACA subsidies recently contributed to a government funding standoff, and CRFB argued that the new Senate bill reflects a political compromise that prioritizes short‑term relief over long‑term fiscal responsibility.
“After a pointless government shutdown over this issue, it is beyond disappointing that this is the preferred solution to such an important issue,” MacGuineas wrote.
The off-year elections cast the government shutdown and cost-of-living arguments in a different light. Democrats made stunning gains and almost flipped a deep-red district in Tennessee as politicians from the far left and center coalesced around “affordability.”
Senate Minority Leader Chuck Schumer is reportedly smelling blood in the water and doubling down on the theme heading into the pivotal midterm elections of 2026. President Donald Trump is scheduled to visit Pennsylvania soon to discuss pocketbook anxieties. But he is repeating predecessor Joe Biden’s habit of dismissing inflation, despite widespread evidence to the contrary.
“We fixed inflation, and we fixed almost everything,” Trump said in a Tuesday cabinet meeting, in which he also dismissed affordability as a “hoax” pushed by Democrats.
Lawmakers on both sides of the aisle now face a politically fraught choice: allow premiums to jump sharply—including in swing states like Pennsylvania where ACA enrollees face double‑digit increases—or pass an expensive subsidy extension that would, as CRFB calculates, explode the deficit without addressing underlying health care costs.
Netflix Inc. has won the heated takeover battle for Warner Bros. Discovery Inc. Now it must convince global antitrust regulators that the deal won’t give it an illegal advantage in the streaming market.
The $72 billion tie-up joins the world’s dominant paid streaming service with one of Hollywood’s most iconic movie studios. It would reshape the market for online video content by combining the No. 1 streaming player with the No. 4 service HBO Max and its blockbuster hits such as Game Of Thrones, Friends, and the DC Universe comics characters franchise.
That could raise red flags for global antitrust regulators over concerns that Netflix would have too much control over the streaming market. The company faces a lengthy Justice Department review and a possible US lawsuit seeking to block the deal if it doesn’t adopt some remedies to get it cleared, analysts said.
“Netflix will have an uphill climb unless it agrees to divest HBO Max as well as additional behavioral commitments — particularly on licensing content,” said Bloomberg Intelligence analyst Jennifer Rie. “The streaming overlap is significant,” she added, saying the argument that “the market should be viewed more broadly is a tough one to win.”
By choosing Netflix, Warner Bros. has jilted another bidder, Paramount Skydance Corp., a move that risks touching off a political battle in Washington. Paramount is backed by the world’s second-richest man, Larry Ellison, and his son, David Ellison, and the company has touted their longstanding close ties to President Donald Trump. Their acquisition of Paramount, which closed in August, has won public praise from Trump.
Comcast Corp. also made a bid for Warner Bros., looking to merge it with its NBCUniversal division.
The Justice Department’s antitrust division, which would review the transaction in the US, could argue that the deal is illegal on its face because the combined market share would put Netflix well over a 30% threshold.
The White House, the Justice Department and Comcast didn’t immediately respond to requests for comment.
US lawmakers from both parties, including Republican Representative Darrell Issa and Democratic Senator Elizabeth Warren have already faulted the transaction — which would create a global streaming giant with 450 million users — as harmful to consumers.
“This deal looks like an anti-monopoly nightmare,” Warren said after the Netflix announcement. Utah Senator Mike Lee, a Republican, said in a social media post earlier this week that a Warner Bros.-Netflix tie-up would raise more serious competition questions “than any transaction I’ve seen in about a decade.”
European Union regulators are also likely to subject the Netflix proposal to an intensive review amid pressure from legislators. In the UK, the deal has already drawn scrutiny before the announcement, with House of Lords member Baroness Luciana Berger pressing the government on how the transaction would impact competition and consumer prices.
The combined company could raise prices and broadly impact “culture, film, cinemas and theater releases,”said Andreas Schwab, a leading member of the European Parliament on competition issues, after the announcement.
Paramount has sought to frame the Netflix deal as a non-starter. “The simple truth is that a deal with Netflix as the buyer likely will never close, due to antitrust and regulatory challenges in the United States and in most jurisdictions abroad,” Paramount’s antitrust lawyers wrote to their counterparts at Warner Bros. on Dec. 1.
Appealing directly to Trump could help Netflix avoid intense antitrust scrutiny, New Street Research’s Blair Levin wrote in a note on Friday. Levin said it’s possible that Trump could come to see the benefit of switching from a pro-Paramount position to a pro-Netflix position. “And if he does so, we believe the DOJ will follow suit,” Levin wrote.
Netflix co-Chief Executive Officer Ted Sarandos had dinner with Trump at the president’s Mar-a-Lago resort in Florida last December, a move other CEOs made after the election in order to win over the administration. In a call with investors Friday morning, Sarandos said that he’s “highly confident in the regulatory process,” contending the deal favors consumers, workers and innovation.
“Our plans here are to work really closely with all the appropriate governments and regulators, but really confident that we’re going to get all the necessary approvals that we need,” he said.
Netflix will likely argue to regulators that other video services such as Google’s YouTube and ByteDance Ltd.’s TikTok should be included in any analysis of the market, which would dramatically shrink the company’s perceived dominance.
The US Federal Communications Commission, which regulates the transfer of broadcast-TV licenses, isn’t expected to play a role in the deal, as neither hold such licenses. Warner Bros. plans to spin off its cable TV division, which includes channels such as CNN, TBS and TNT, before the sale.
Even if antitrust reviews just focus on streaming, Netflix believes it will ultimately prevail, pointing to Amazon.com Inc.’s Prime and Walt Disney Co. as other major competitors, according to people familiar with the company’s thinking.
Netflix is expected to argue that more than 75% of HBO Max subscribers already subscribe to Netflix, making them complementary offerings rather than competitors, said the people, who asked not to be named discussing confidential deliberations. The company is expected to make the case that reducing its content costs through owning Warner Bros., eliminating redundant back-end technology and bundling Netflix with Max will yield lower prices.
Nearly all leading artificial intelligence developers are focused on building AI models that mimic the way humans reason, but new research shows these cutting-edge systems can be far more energy intensive, adding to concerns about AI’s strain on power grids.
AI reasoning models used 30 times more power on average to respond to 1,000 written prompts than alternatives without this reasoning capability or which had it disabled, according to a study released Thursday. The work was carried out by the AI Energy Score project, led by Hugging Face research scientist Sasha Luccioni and Salesforce Inc. head of AI sustainability Boris Gamazaychikov.
The researchers evaluated 40 open, freely available AI models, including software from OpenAI, Alphabet Inc.’s Google and Microsoft Corp. Some models were found to have a much wider disparity in energy consumption, including one from Chinese upstart DeepSeek. A slimmed-down version of DeepSeek’s R1 model used just 50 watt hours to respond to the prompts when reasoning was turned off, or about as much power as is needed to run a 50 watt lightbulb for an hour. With the reasoning feature enabled, the same model required 7,626 watt hours to complete the tasks.
The soaring energy needs of AI have increasingly come under scrutiny. As tech companies race to build more and bigger data centers to support AI, industry watchers have raised concerns about straining power grids and raising energy costs for consumers. A Bloomberg investigation in September found that wholesale electricity prices rose as much as 267% over the past five years in areas near data centers. There are also environmental drawbacks, as Microsoft, Google and Amazon.com Inc. have previously acknowledged the data center buildout could complicate their long-term climate objectives.
More than a year ago, OpenAI released its first reasoning model, called o1. Where its prior software replied almost instantly to queries, o1 spent more time computing an answer before responding. Many other AI companies have since released similar systems, with the goal of solving more complex multistep problems for fields like science, math and coding.
Though reasoning systems have quickly become the industry norm for carrying out more complicated tasks, there has been little research into their energy demands. Much of the increase in power consumption is due to reasoning models generating much more text when responding, the researchers said.
The new report aims to better understand how AI energy needs are evolving, Luccioni said. She also hopes it helps people better understand that there are different types of AI models suited to different actions. Not every query requires tapping the most computationally intensive AI reasoning systems.
“We should be smarter about the way that we use AI,” Luccioni said. “Choosing the right model for the right task is important.”
To test the difference in power use, the researchers ran all the models on the same computer hardware. They used the same prompts for each, ranging from simple questions — such as asking which team won the Super Bowl in a particular year — to more complex math problems. They also used a software tool called CodeCarbon to track how much energy was being consumed in real time.
The results varied considerably. The researchers found one of Microsoft’s Phi 4 reasoning models used 9,462 watt hours with reasoning turned on, compared with about 18 watt hours with it off. OpenAI’s largest gpt-oss model, meanwhile, had a less stark difference. It used 8,504 watt hours with reasoning on the most computationally intensive “high” setting and 5,313 watt hours with the setting turned down to “low.”
OpenAI, Microsoft, Google and DeepSeek did not immediately respond to a request for comment.
Google released internal research in August that estimated the median text prompt for its Gemini AI service used 0.24 watt-hours of energy, roughly equal to watching TV for less than nine seconds. Google said that figure was “substantially lower than many public estimates.”
Much of the discussion about AI power consumption has focused on large-scale facilities set up to train artificial intelligence systems. Increasingly, however, tech firms are shifting more resources to inference, or the process of running AI systems after they’ve been trained. The push toward reasoning models is a big piece of that as these systems are more reliant on inference.
Recently, some tech leaders have acknowledged that AI’s power draw needs to be reckoned with. Microsoft CEO Satya Nadella said the industry must earn the “social permission to consume energy” for AI data centers in a November interview. To do that, he argued tech must use AI to do good and foster broad economic growth.
Politics8 years ago