Technology

➚ MOVE FAST: Signal. The encrypted messaging app at the center of the White House national security leak couldn’t have asked for a better marketing campaign. It could boost donations to support the service, which is owned by a nonprofit started by WhatsApp co-founder Brian Acton.

➘ BREAK THINGS: Noise. The Trump administration launched its first major effort to curb Chinese tech advancements, with more than 50 companies in the People’s Republic added to the US export blacklist. But such moves seem increasingly less effective, with Chinese startups like DeepSeek making significant progress even with restrictions in place.

Toxic relationship. Last month, we reported on people who have romantic relationships with chatbots, including ChatGPT. Their usage varied, with some utilizing their chatbot companion as a friendly sounding board and others forming deep emotional connections, even considering themselves married to the chatbot. In one of the first studies of its kind, OpenAI and the MIT Media Lab investigated how individuals interact emotionally with ChatGPT and how that impacts their feelings of real-world loneliness.

Using ChatGPT’s voice feature in a limited capacity helped mitigate feelings of loneliness, the study found. However, increased daily usage “correlated with higher loneliness, dependence, and problematic use, and lower socialization.”

Notably, women are at a higher risk of distancing themselves from real people after prolonged chatbot usage than men, in part because they are more likely to use the chatbot for personal discussions, the study found. Meanwhile, men more so engage in practical, task-oriented conversations. For both men and women, when the gender of the chatbot’s voice was opposite their own, they reported higher levels of loneliness and emotional dependence.

With loneliness levels reaching new highs, technology has cropped up as a potential solution. While more research is needed on the impact of AI on mental health, the study determined that “focusing solely on technical solutions is insufficient” in addressing the loneliness epidemic. Establishing guardrails and AI policy is important, but “the broader issue lies in ensuring people have strong social support systems in real life.”

There are now roughly 600,000 components inside a data center rack, or high-powered computers stacked as tall as an NBA point guard, Nvidia CEO Jensen Huang revealed last week.

The blazing-fast chips, liquid cooling gear, fiber-optic cables and other parts are packed so densely that engineers aren’t sure whether the floors will start to slump under the weight. Each piece of equipment and software, which fill buildings more than an acre large, represents a business opportunity for companies hoping to ride the wave of unprecedented investment in artificial intelligence.

But the success and failure of companies betting on the AI boom may depend on whether it’s public or private market investors bankrolling them.

Jon McNeill, former president of Tesla and chief operating officer of Lyft, is both. McNeill, who has seen the AI industry up close in various roles in the tech industry, launched a company called VistaShares that offers exchange-traded funds tracking various tech verticals, including AI infrastructure.

The GM board member is also an investor in private tech companies, and says the biggest difference is that the scope of AI infrastructure development still hasn’t sunk in for investors in public stocks. That’s how he explains recent market gyrations, like the $1 trillion wipeout in market value of tech stocks after Chinese startup DeepSeek revealed its latest AI advancements.

“I’ve had investors say ‘Jesus, how long is the runway to invest in these foundation models?’” McNeill said in an interview with Semafor.

His answer: decades. “Look at the internet super cycle that started in 1997,” McNeill said. “We’re 27 years in, and there has not been a let up. AI infrastructure is like the internet.”

The World Economy Summit 2025 will bring together US Cabinet officials, global finance ministers, central bankers, and over 200 CEOs of the world’s largest companies. The three-day summit will take place Apr. 23-25, 2025, in Washington, DC, and will be the first of its kind since the new US administration took office. Featuring on-the-record conversations with top executives such as Alex Chriss, President and CEO, PayPal; Adena Friedman, Chair and CEO, Nasdaq; Reid Hoffman, Co-Founder, Manas AI & Partner, Greylock; Ted Sarandos, Co-CEO, Netflix; and Evan Spiegel, CEO, Snap, the summit will advance dialogues that catalyze global growth and fortify resilience in an uncertain, shifting global economy.

Apr. 23-25 | Washington, DC | Learn More

The number of years until quantum computing tech will be able to run the kind of practical applications that today’s computers can’t, Google quantum executive Julian Kelly told CNBC, echoing what CEO Sundar Pichai told Semafor in December. The estimation is on the optimistic side, contrasting Nvidia chief Jensen Huang’s statement earlier this year that practical quantum computing would take two decades.

Generative AI is so last year. Companies are launching agentic AI platforms — consisting of a web of bots that communicate with each other to perform tasks — in an effort to free up human workers and increase efficiency. But these agents are sometimes working across hundreds of internal systems without a high threshold of data quality or homogeneity. That can present a bit of a nesting doll problem — agents relying on more deeply layered agents that can get stuck, return inaccurate results, or loop requests back and forth like a snake eating its own tail.

The $5.7 billion data management company Informatica is taking on inter-agent communication as its latest AI project. “One piece of software can become 100 agents that you can multiply across the tech stack,” CEO Amit Walia told Semafor. “That actually creates a massive problem for our customers, because each agent requires a harmony of data. There has to be a handoff from agent ‘A’ to agent ‘B’ to do an activity.”

Informatica’s agentic platform, coming later this year, will homogenize data across its customers’ internal interfaces so their agents can work smarter, Walia said.

The company, which has been around for 30 years, is stuck in a weird spot between the flashy startups and resource-heavy Big Tech giants. Its stock hasn’t reaped the same benefits of Wall Street’s tech shopping spree, with the latest hit coming from disappointing quarterly earnings on the back of fewer customer renewals, though year-over-year profits are largely up. Walia doesn’t expect its agentic platform to be profitable for a few years.

The issue, in part, is that Informatica is waiting on other companies to implement agentic AI on a greater scale, he said. The company’s agents can’t do much until its customers deploy theirs — something Walia expects in late 2025 to early 2026.

Safeguard Treasury bonds. Secure the cloud. Undermine China’s race for influence across the global South.

Companies across disparate sectors — cryptocurrency, infrastructure, cybersecurity, and energy — are hitching their agendas to the Trump administration’s, Semafor’s Rohan Goswami and Liz Hoffman reported Tuesday. Deals of all stripes are being recast as national security imperatives, which executives hope will steer them past a mercurial White House whose regulatory priorities remain unclear.

Take Tether: The crypto giant is on a charm tour in Washington, where it is pushing stablecoin legislation and trying to assuage concerns that its products are used by criminals and terrorists. CEO Paolo Ardoino has pitched his company’s product as a way to keep America’s ballooning debt out of unfriendly or unstable hands. Tether said it is the 7th-largest buyer in the world of Treasury bonds, which it holds as collateral backing its coins.

Google framed its $32 billion acquisition of Wiz, which needs antitrust approval, as literally helping to safeguard national security. “Organizations of all sizes — from startups and large enterprises to governments and public sector organizations — can use Wiz to protect” against cyberattacks, Google wrote in its deal announcement. It was a not-so-subtle nudge to staffers at Treasury, State, and Commerce — all of which have been infiltrated by Chinese hackers.

There is now a price tag on the deluge of deals the United Arab Emirates plans in the US, Semafor’s Kelsey Warner reports.

The UAE has committed $1.4 trillion over the next decade to “substantially increase… existing investments” in artificial-intelligence infrastructure, semiconductors, energy, and manufacturing, according to a White House statement. The agreement follows UAE National Security Advisor Sheikh Tahnoon bin Zayed’s visit to Washington.

The UAE has already deployed more than $1 trillion in the world’s largest economy, from a stake in Chicago’s parking meters to a majority of Nasdaq-listed chipmaker GlobalFoundries.