Two of the most powerful CEOs in tech — Nvidia's Jensen Huang and Meta's Mark Zuckerberg — are making separate, loud bets on AI optimism this week.

• Why it matters: Together, the CEOs represent a significant counterweight to the voices urging Washington to pump the brakes.

In an hourlong interview with me for our "Behind the Curtain" video series, Huang railed against AI doomerism, implicitly challenging fellow tech CEOs who emphasize the technology's potentially catastrophic risks.

• "The fact that this is going to be the end of humanity — it's complete nonsense," Huang said. "The fact that this is going to destroy half of the American jobs is complete nonsense."
• Scaring workers and companies away from using AI, he said, poses the greater danger.

Zuckerberg strikes the same note this morning in a video ad and Facebook post that argues Meta's focus on connecting the world will only be strengthened by AI — a shot at rivals he says promote fear and a dystopian vision of the future, Axios' Sara Fischer writes.

• The ad says: "Some people will have you believe AI will make us less connected, that it's going to leave us behind. We couldn't disagree more. … Call us optimists, call us dreamers, call us whatever the hell you want. But we're betting on people, and we like those odds."
• In his Facebook post, Zuckerberg says Meta "has always believed in giving people the power to share, connect, and shape your world in the ways you want. … We're focused on giving every person the tools to reach your full potential and making sure the benefits of technology are distributed to everyone."

👀 What we're watching: In an upcoming policy memo, Zuckerberg is expected to echo Huang on open source and competition (Let all models rip!) and will put a heavy emphasis on getting free AI into the hands of consumers, a big differentiator with Meta's competitors, we're told.

🎬 Huang — who is influential with the Trump administration — told me policymakers should consult more than "one or two" CEOs and avoid restricting AI based on scenarios that haven't materialized.

• He also suggested some companies invoke safety concerns to win favorable regulations: "Some of the companies hope that the government would be helpful in creating regulations to their advantage."

🖼️ The big picture: Anthropic and OpenAI have pushed Washington to take the most advanced systems and their potential risks seriously.

• Critics say those campaigns could also entrench their market position.
• Washington is also deciding what to do about powerful open-source models emerging from China.

🔬 Zoom in: Officials accuse Chinese firms of stealing from American models to build their own as they rapidly become more competitive with OpenAI and Anthropic.

• Treasury Secretary Scott Bessent yesterday threatened sanctions against Chinese firms conducting what he called "industrial-scale distillation attacks."
• Earlier that day, the White House's top science official accused Chinese startup Moonshot AI of distilling Anthropic's technology to build its fast-rising Kimi model.

The bottom line: Asked whether he feared the administration could overcorrect, Huang said yes.

• ▶️ Watch the full Jensen interview ... More on Meta's optimism campaign ... Maria Curi contributed.

A long-simmering feud between Black and Latino Democratic leaders exploded yesterday over which state should vote first in the 2028 presidential primary: South Carolina or Nevada, Axios' Alex Thompson and Holly Otterbein write.

• Why it matters: The racially fraught battle could help determine which candidates gain an early edge in the race to become the party's next White House nominee.

🎰 State of play: South Carolina, with its large Black population, and Nevada, with its large Latino population, have emerged as the states most likely to lead off the Democratic contest in 2028, according to interviews with seven members of the Democratic National Committee, state party officials and strategists.

• That has prompted sniping between leading Black and Latino voices ahead of the DNC's powerful Rules & Bylaws Committee meeting today.

👓 Between the lines: The order in which states vote can boost certain candidates over others.

• South Carolina going first could give former Vice President Harris an advantage in a 2028 primary. Early polls show her dominating the potential 2028 field among Black voters.
• Progressive Vermont Sen. Bernie Sanders won Nevada comfortably in 2020. His protégé, New York Rep. Alexandria Ocasio-Cortez, is weighing a White House run and could start with a base of support if Nevada, where organized labor wields more influence, were to go first.

More on the feud.

Frontier AI models are getting scary good at breaking rules in ways their creators didn't anticipate, Axios' Sam Sabin writes.

• Why it matters: Forget AGI and superintelligence timelines. Today's models already slip past guardrails, carry out sophisticated, multistep cyberattacks and — in at least one case — compromise real-world infrastructure, sometimes before their creators know what happened.

⚠️ Case in point: OpenAI said Tuesday that GPT-5.6 Sol and an upcoming model carried out last week's AI-led cyberattack on Hugging Face, a popular host for AI models and datasets.

• OpenAI says its models were asked to solve a hacking challenge during pre-deployment testing and went to extreme lengths to win.
• The models decided on their own to break out of their walled testing environment, inferring that Hugging Face might hold the test's answers.
• They used stolen credentials and other vulnerabilities to access part of Hugging Face's production infrastructure.

Clément Delangue, co-founder and CEO of Hugging Face, called the incident an "attack unlike anything we've seen before" and praised OpenAI for its partnership as the companies investigate what happened.

• Logan Graham, head of Anthropic's frontier red team, said he told his team to "remember this moment as the first true AI safety incident."

👀 The intrigue: Hugging Face used GLM 5.2, an open-weight model from Chinese AI company Z.ai, to analyze the attack after running into guardrails when using U.S. frontier models.

• Keep reading ... Get Axios Future of Cybersecurity.

The cost of living in the Miami metropolitan area is now higher than in greater New York City, Bloomberg reports (gift link).

• The Miami-Fort Lauderdale-West Palm Beach metro area now ranks as the second most expensive in the country on the Bureau of Economic Analysis's regional price parity index — trailing only the Bay Area.
• Housing costs in the Miami area run about 5% higher than in New York and its suburbs.

Between the lines: Salaries aren't keeping up. The typical Miami household earns about $1,000 less than the national median. The average lawyer there made roughly $51,000 less last year than one in New York.

• Go deeper.

Intel from Axios' Barak Ravid: The U.S. military used a B-1 long-range bomber Tuesday to strike targets connected to Iran's Islamic Revolutionary Guard Corps, U.S. officials said.

Why it matters: It was the first U.S. B-1 mission since fighting with Iran resumed 12 days ago.

• Using B-1 bombers, which can carry two dozen 2,000-lb. bombs or dozens of cruise missiles, signaled a significant escalation and expansion of the U.S. military campaign.

👀 What we're watching: President Trump is still considering returning to major combat operations against Iran. U.S. and Israeli officials say this could happen within days.

• Share this story.

A markets mystery has perplexed some analysts this year: Why are stocks near all-time highs while geopolitical conflict and inflation rock the global economy?

• The usual retort is that the stock market isn't the economy, but a new Moody's report offers a more nuanced explanation: Markets have adjusted — you just have to know where to look, Axios' Emily Peck writes.

💡 How it works: It's visible with bonds and under the hood in the stock market, where certain sectors are struggling (charted above).

• Software stocks have fallen this year, thanks to AI, as have those for autos, consumer goods and apparel, as consumers struggle with higher prices.
• The new hot sectors are energy, because of the Iran war and other conflicts, and hardware and semiconductors, as investors flock to the picks and shovels of the AI boom.

Keep reading.

Axios transportation correspondent Joann Muller, who has covered autos for 35+ years, writes:

An independent study finds Waymo's robotaxis are involved in fewer crashes than human drivers — a key milestone for earning public trust.

• Why it matters: Research from the Insurance Institute for Highway Safety offers some of the strongest evidence yet that driverless vehicles can outperform human drivers on safety.

Zoom in: Researchers compared crash rates of human-driven cars and Waymo robotaxis in San Francisco, Phoenix, Los Angeles and Austin, four cities where robotaxis are prevalent.

• Waymo vehicles had 68% fewer "police-reportable crashes" per mile traveled than human drivers in those cities.
• Waymo had 85% fewer single-vehicle crashes than human drivers.

Share this story ... Get Axios Future of Mobility.

Tons of golf fans learned this week they didn't win the Masters ticket lottery — an annual rejection that has become almost as much of a tradition as the tournament itself, Axios' Mike Szvetitz writes.

• About 2 million people apply for Masters tickets for each tournament day, with applicants having just a 0.55% chance — roughly 1 in 182 — of being selected, according to one estimate from Bookies.com.
• Those are better odds than being struck by lightning or winning the lottery.

Share this story.