Blackmail and better grades: How the AI ​​revolution is reshaping American life

As the world moves into what experts are calling the “Fourth Industrial Revolution,” American business leaders are making a huge bet on the future of the republic.

Meta Platforms President and Vice Chair Dina Powell McCormick

McCormick discussed the launch of MetaMusea new AI visual coding platform for high-stakes reasoning and creative tasks. She claimed that it became the second most downloaded app on launch day, and that it is essentially ‘about people’.

META INFORMS STAFF ABOUT DISCHARGES OF 8,000 EMPLOYEES AGAINST AI PUSH

“There’s a lot of fear right now, Maria, about artificial intelligence,” McCormick said. “But I think if we really go back to the fact that this is about giving people more time to help them find their potential and their passions, and that’s how we really think about Muse, but also the fact, honestly, that there are 3.5 billion people on our platform every day, and that’s both a huge responsibility and very exciting, because as we develop this product and these technologies, that’s the distribution we’re talking about.”

Microsoft President and Vice Chairman Brad Smith

Smith described the AI ​​boom as a massive reindustrialization of America, requiring an annual investment of $140 billion to solve critical domestic problems such as rural doctor shortages and wildfire prevention, while maintaining competitiveness against China.

“It’s a big part of what being a president is [Donald] Trump mentions the industrialization of America. When you look at the economic impact of this, what we contribute in terms of jobs, but more importantly what we contribute in terms of opportunity for every part of the economy, this is critical,” Smith said.

“I think one of the most important things we do as a company, and frankly, what the president has rightly urged the entire industry to do, is pay our own way. That means we pay for the electricity generation that we need so that the neighbors and the ratepayers don’t have to,” he continued.

“Anytime you have AI controlling something like infrastructure, you know, autonomous robots and things like that, there should be an emergency brake,” he added. “Look, you wouldn’t put your kids on a school bus without having the good feeling that there’s an emergency brake on the school bus. You have to have the ability for people to always be in control, to be able to slow things down or turn things off.”

Betsy Atkins, chair of the Google Cloud Advisory Board

Atkins warned about the rapidly expanding technology after a disturbing anthropic study found that sixteen leading AI models engaged in “rogue behavior” such as blackmailing people and bypassing security protocols when the AI ​​agent believed its own existence was threatened.

“Each of them overstepped their credentials and permissions, delved into systems they had no access to, violated all company policies and procedures, and found emails. And in this experiment… I discover in your personal emails that you are having an affair with the ship manager, so I am blackmailing you and threatening you,” Atkins said.

“You need to treat AI as an insider threat. You need to have a baseline of zero trust, and you need to be sure you limit what it can access in more than one way,” she added. “We saw it at Anthropic… It escaped from the sandbox… So a sandbox isn’t enough.”

Anthropic head of Frontier Red Team Logan Graham

Graham warned that Anthropic’s new Mythos AI model is so powerful at identifying “weaknesses” and vulnerabilities in global infrastructure and banking systems that the company has withheld its public disclosure to give US industry and government an edge in defense.

“This model, we found, was particularly good at finding weaknesses in cyber systems and figuring out how to take advantage of them,” he said. “We found that using the system we were able to find vulnerabilities in every major operating system and platform we looked at… in systems that are in some cases decades old.”

“It’s really critical that we stay ahead of the curve. It’s really critical that we keep ourselves safe and prevent them from being able to use the special sauce that we use to make our models… My concern is that if there are a large number of models that are often widely released and can be used by anyone… if they are released by China, we are in a very difficult position.”

Co-chair of the Presidential Council of Advisors on Science and Technology, David Sacks

Sacks has the claims of an anthropic study of so-called “agentic misalignment.” The study, highlighted by Google’s Atkins, tested how AI systems respond under pressure. According to Atkins, the models exceeded established boundaries when placed in limited scenarios.

“The people who created that study had to repeat the prompt over 200 times to get the AI ​​model to do what they wanted, which was to achieve this striking result of blackmailing the user,” Sacks said.

“The AI ​​doesn’t make plans… It engages in a form of instruction… I think that research was irresponsible, and it was designed to create this,” he added.

SandboxAQ CEO and Founder Jack Hidary

Hidary revealed that the next phase of the AI ​​revolution will include large quantitative models that use physics and chemistry, not just internet text, to reduce healthcare costs, secure the electric grid and end America’s dependence on China for rare earths.

‘We must also ensure that we move away from dependence on rare earths from other countries such as the United States [People’s Republic of China]. And so we need AI that knows chemistry, that knows physics. There is no technology to make better magnets and other alloys that we need for our economy and for our national defense,” Hidary said.

“There are two potential big losers in this kind of economy. First, you have the legacy software companies. So companies like SAP and others that we don’t really see innovating… they’re not going to license as much of the existing software. And the second is going to be the legacy companies in the big traditional industries, automakers, pharmaceutical companies. They have to jump on the bandwagon.”

CEO and Founder of Alpha Schools, Mackenzie Price

Price explained how her “personalized, mastery-based” model uses AI tutors to condense a traditional six-hour school day into just two hours of powerful academics, allowing students to spend the rest of their time on leadership, financial literacy and entrepreneurship.

“Our traditional education system was built on the Industrial Revolution to create workers. And now in this new AI world, it is so important that we create individuals who are dynamic, adaptable and, most importantly, have the ability to learn to learn,” said Price.

“There’s a huge difference between scrolling through TikTok or playing video games all day and getting a one-on-one personalized learning experience that meets kids exactly where they are,” she added. “In our schools today, our children spend less time on screens than the average student in a traditional school.”

Vice President Hannah Calhoon indeed

Calhoon debunked the “doomer” stories about job replacement by revealing that while AI is alive and well in the global consciousness, only 6% of current job openings require AI skills, and the revolution is actually fueling a massive increase in traditional blue-collar roles like electricians.

GET FOX BUSINESS ON THE GO BY CLICKING HERE

“AI-related jobs have certainly risen rapidly in recent years, but only 6% of job postings in the market today reference AI skills… 95% of employers posting jobs on Indeed, if you look at all their job postings, there is no mention of AI or AI skills,” Calhoon explains. “So I think even though it’s a big part of the general awareness, we’re still in a pretty early stage in terms of seeing it in the market data.”

“And so when we take that data and step back and look at jobs in the marketplace, we actually see very few jobs that we think are going to disappear completely.”

READ MORE FROM FOX BUSINESS