Good morning. Andrew here. We’ve got a probable new Fed governor — at least on a temporary basis. We dive into what President Trump’s pick means for interest rates, Jay Powell and more. Meanwhile, OpenAI released its latest A.I. model, upgrading ChatGPT along the way. (I’ve been playing with it and it is quite something — try it if you can.) And speaking of A.I., we’ve got some creative hacks that a financial C.E.O. has been using. (Was this newsletter forwarded to you? Sign up here.)
Trump’s new man at the FedPresident Trump’s announcement that he would nominate Stephen Miran, a former hedge fund executive turned top economic adviser, to temporarily fill a vacancy at the Fed puts the White House a step closer to reshaping the central bank. If confirmed by the Senate, Miran would get a key vote on interest-rate policy, putting more pressure on the current Fed chair, Jay Powell, to lower borrowing costs even as concerns grow about the inflationary effects of Trump’s tariffs. The market response to Miran was swift. The dollar fell again this morning against a basket of currencies. His nomination “raises another red flag of risk on U.S. assets” and could rekindle concerns about the so-called sell America trade, Richard Cochinos, a foreign exchange strategist at RBC Capital Markets, wrote in a research note yesterday. And futures traders now anticipate two or three rate cuts this year, with the first coming at the next meeting in September. What’s worrying investors:
The appointment increases pressure on the Fed’s political independence. The central bank’s relative freedom from government interference is seen as a kind of North Star for global investors. It was reinforced this spring by the Supreme Court, in the face of Trump’s repeated grumbling that he would like to fire Powell. But the chessboard maneuvering at the Fed is complicated. Miran may be on the board of governors for only a few months: Trump has made clear that he sees Miran’s appointment as temporary. What Wall Street remains focused on is what will happen to Powell, whose term as chair is up in May (but who legally can remain a governor until 2028). Trump is drawing up a short list to replace him, with Christopher Waller, an existing Fed governor and Trump appointee who has voted for lower rates, emerging as a front-runner, Bloomberg reports. Others in contention include two Kevins: Hassett, the director of the National Economic Council, and Warsh, a former Fed official.
President Trump’s trade war grows. Traders were surprised to learn that U.S. Customs and Border Protection has imposed new levies on the import of gold bars, The Financial Times reports, the latest blow to Switzerland. Brazil and India, who also have been unable to evade bruising tariffs, have vowed to strengthen their trade ties and cooperate more on defense. Trump potentially opens U.S. retirement accounts to crypto and private equity investments. The president signed an executive order yesterday directing the Labor Department to re-evaluate whether Wall Street and the crypto community could get a bigger piece of the $12.2 trillion held in 401(k) accounts. Bitcoin briefly rallied on the news. In a separate executive order, Trump renewed his campaign to bar “debanking,” the claim that lenders are refusing to do business with conservatives over their political views. Trump calls on Intel’s C.E.O. to step down over allegations of conflicts of interest. “The CEO of INTEL is highly CONFLICTED and must resign, immediately,” Trump wrote on social media yesterday of Lip-Bu Tan, adding, “There is no other solution to this problem.” The president’s post didn’t elaborate, but came soon after Senator Tom Cotton, Republican of Arkansas, asked Intel’s chair about Tan’s previous business ties to China. (The Wall Street Journal adds that Tan has clashed with the chipmaker’s board over strategy.) What’s next for OpenAIA week of big news for OpenAI — including our scoop about its latest fund-raising round and the introduction of its open-source artificial intelligence models — wrapped up yesterday with the unveiling of GPT-5, the company’s flagship A.I. model. All are signs that OpenAI, which may soon become the world’s most valuable privately held start-up, is succeeding wildly. But even these developments underscore the huge challenges that the A.I. juggernaut faces. Point No. 1: GPT-5. OpenAI says the software, which will underpin both free and paid versions of ChatGPT, is better than its predecessors at complex tasks like coding and in feeling “more human.” The advances will probably solidify ChatGPT’s dominance among consumers. But the bigger question is whether they will help OpenAI amass paid corporate subscriptions, a sector where the company faces stiff competition from Anthropic and Google — and is an arena in which companies tend to rely on a mix of A.I. providers. Point No. 2: new open-source software. Earlier this week, OpenAI announced two A.I. models that are freely available for public use. They are meant to compete against rival open-source offerings from Meta and China’s DeepSeek. OpenAI’s new models aren’t as good as GPT-5. But the risk is that they’re good enough for many companies to use, foregoing expensive enterprise subscriptions — even as OpenAI hopes heavy-duty users will still pay up for the bleeding-edge stuff. Point No. 3: OpenAI’s soaring valuation. The start-up is in talks with the venture capital firm Thrive Capital and other investors to let them buy shares from current and former OpenAI employees, valuing the company at about $500 billion, DealBook hears. (The negotiations were reported earlier by Bloomberg.)
That’s up sharply from the $300 billion valuation that OpenAI negotiated with SoftBank and others earlier this year. But the ever-climbing valuation also puts more pressure on OpenAI to overhaul its corporate structure to become a for-profit company — a process that needs the assent of Microsoft, a key tech partner — since that would unlock the path to eventually going public.
UnitedHealth deal gets the green lightTwo years after announcing a $3.3 billion deal to expand its reach into home health and hospice care, UnitedHealth Group has reached a proposed settlement with the Justice Department that allows it to buy Amedisys, a large home-health company, as long as the companies divest dozens of facilities as part of the merger, Reed Abelson reports. The Justice Department’s response to the merger was closely watched by lawmakers and regulators, many of whom cited the Amedisys deal as more evidence that UnitedHealth had grown too big, potentially hampering competition. UnitedHealth, whose operations span health care, had $400 billion in revenue in 2024. The details: Under the proposed settlement, which must still be approved by a federal judge, the two companies would divest 164 home health and hospice locations across 19 states, estimated to account for roughly half a billion dollars in annual revenue. The buyers, according to court filings, would be BrightSpring Health Services, a private-equity backed firm, and the Pennant Group. Amedisys will also pay a $1.1 million penalty related to its response to a request for documents from the Justice Department. UnitedHealth isn’t completely off the hook. It was unclear how vigorously the Trump administration’s Justice Department would pursue the antitrust lawsuit to block the acquisition, which was filed under Joe Biden last November. Though the health care conglomerate is being allowed to complete the deal, it continues to be under widespread scrutiny. The company recently announced it was responding to both civil and criminal inquiries from the Justice Department regarding its Medicare business. The D.O.J. said forcing the divestment of facilities addressed the concerns. “This settlement protects quality and price competition for hundreds of thousands of vulnerable patients and wage competition for thousands of nurses,” said Gail Slater, the Justice Department’s antitrust chief. The department did not respond to a request for additional comment.
Talking A.I. with the C.E.O. of Principal FinancialEvery week, we’re asking a C.E.O. how he or she uses generative artificial intelligence. Deanna Strable, who leads the investment management and insurance company Principal Financial, says she uses it for performance reviews and that the company recently introduced A.I. training for its 20,000 employees. Her responses have been edited and condensed for clarity. How do you use A.I. personally? I’m planning a trip with five of my friends, and I asked a chatbot to put together a four-day itinerary, given our age, activities that we like to do. We’re also getting a lot of traction with using A.I. to help with performance management: After I get feedback from a lot of people, I can use A.I. to compare it to prior feedback or highlight things that have improved. What direction have you given your team on how to use A.I.? It’s changing how I lead. I’ve been training myself to ask leaders who come in with a problem if they’ve explored how A.I. might help solve it differently. The first couple times, you get blank stares. But as you continue to do that, they become more inquisitive and ultimately they’re embedding it in how they’re leading their organization. If I can embed it into how I’m interacting with my people, it’ll start influencing down throughout the organization. We’re also training all 20,000 of our employees in A.I. fluency and data literacy. You can’t just allow people who are curious about it to drive it. Ultimately, it needs to go through the organization. The leaders of each function will be able to decide how to use enhanced productivity: Do they drive more throughput? Do the same level of throughput with less people? It’ll probably be somewhere in between. We always have more ideas than we have the capability to do. We hope you’ve enjoyed this newsletter, which is made possible through subscriber support. Subscribe to The New York Times.
Deals
Politics, policy and regulation
Best of the rest
|
