Turns out 2025 was not the year we bowed before our AI overlords

It’s probably worth noting that Geoffrey Hinton, a researcher and academic widely recognized as “the godfather of AI”, walked away from his position as head of Google’s artificial intelligence research team a couple of years ago in part so he could speak freely about his concerns over the advancing technology.

Hinton, 78, a British-Canadian computer scientist, has spent his life developing machine language and digital neural networks systems which attempt to replicate the workings of the human brain. His work has helped catapult the advancement of AI tools and, since leaving Google, he has been more forthright in sharing his personal concerns over the technology’s potential outcomes.

“The problem is, once these things get more intelligent than us, it’s not clear we’re going to be able to control it,” Hinton said in an interview shortly after resigning from his job at Google. “There are very few examples of more intelligent things controlled by less intelligent things.”

He’s also weighed in on the potential for an AI system to potentially wipe out the human species.

“It’s not inconceivable,” Hinton said. “That’s all I’ll say.”

While the past year has provided no further hints that an AI-driven calamity is imminent, developers released a series of system advancements in 2025 and the effort to integrate AI with almost every digital process continued to accelerate.

Here are some of the biggest AI stories of the year gone by.

Rise of the AI agent

Major developers of AI platforms all released updates to their various toolkits in 2025, advancing the performance of chatbots that respond to inquiries and instructions in natural language, prompt-driven generators that can create videos and still images and integrations into existing and newly created software tools that take advantage of AI’s capabilities.

But the biggest move on the AI front in 2025 may have been the ascension of so-called AI agents from beta status to full-fledged operations. An AI-powered agent is software that can perform tasks on behalf of a user or system. At a basic level, it’s the power of AI to actually operate computer functions or software independent of a user.

Chances are good you have already interacted with an AI agent via an online customer service chatbot, though its potential applications are far broader.

A McKinsey & Co. report released earlier this year notes a few examples of areas that are making use of emerging AI agent capabilities, including:

• Co-pilot agents: These agents serve as copilots for individual users, with the intention of augmenting that person’s productivity and capabilities. Copilot agents, such as Microsoft’s 365 Copilot and OpenAI’s ChatGPT, can assist with drafting content, writing code, or retrieving knowledge. In some cases, copilot agents can act as “smart” assistants that are tailored to a user’s specific workflow.
• Workflow automation platforms: This type of agent focuses on automating single or multistep tasks or smaller workflows, serving as an AI-powered process orchestrator and executor for existing workflows.
• AI virtual workers: AI virtual workers are agents that function as employees or team members and represent the most potentially disruptive category of agents. These virtual workers could enable companies to sidestep full organizational transformation by allowing AI to operate within the company’s current model, which might help capture value more quickly.

In a world that’s growing increasingly wary of differentiating AI content and processes from human creations and interactions, McKinsey researchers note that establishing trust with customers and the general public will be key in advancing AI-driven tools in the business world.

“The companies that derive the most value from AI will be those that create trust with their customers, employees and stakeholders,” said McKinsey partner Roger Roberts in the report. “People must trust AI enough to hand over tasks. Companies’ ethical decisions must be rooted in the values unique to each organization and the values of a society that places humans at the center of the AI ecosystem.”

AI as a profit machine

It’s hard to overestimate how big an impact the artificial intelligence industry had on U.S. investment markets in 2025, which saw overall gains of about 17% in the past year despite a slew of economic challenges.

By one measure, more than 90% of economic growth in the first half of 2025 came from investments in computer equipment and software, which economists chalk up to projects linked to the rush to build data centers and remain in the AI race, per a report from the New York Times. Seven of the top 10 performing stocks in the S&P 500 were lifted by the bet on artificial intelligence and just one of those companies, Nvidia, whose semiconductors have underpinned the AI boom, saw a 36% rise adding some $1 trillion to its market value.

“Equity markets are ending the year on a high note, with the S&P 500 on track for its third consecutive year of double-digit returns, driven by AI momentum and a resilient economy that has shrugged off fiscal and political headwinds,” Craig Johnson, chief market technician at Piper Sandler, said in a note, according to a CNN report.

Billions of dollars in AI-centric investment, supporting everything from product development to the construction of expansive data centers needed for the technology’s growth helped drive new, sky-high valuations, and stock prices, for companies focused on AI efforts. But those speedy ascensions have also raised concerns over so-called bubble conditions that have led to previous market crashes.

And those urging caution even include some of the foremost leaders in the AI race.

Back in August, Sam Altman, CEO of OpenAI, the company behind the wildly popular ChatGPT chatbot, DALL-E image generator and Sora video generator, mused about the state of the AI investment markets.

“Are we in a phase where investors as a whole are overexcited about AI?” Altman asked, according to a report from MIT Technology Review. “My opinion is yes.”

Altman went on to compare the current state of AI investment with the dot-com bubble of the late ‘90s.

“When bubbles happen, smart people get overexcited about a kernel of truth,” Altman said. “Tech was really important. The internet was a really big deal. People got overexcited.”

Google CEO Sundar Pichai concurred with Altman’s assessment in a BBC interview earlier this month, per the MIT report, that there’s “some irrationality” in the current boom. Asked whether Google would be immune to a bubble bursting, Pichai warned, “I think no company is going to be immune, including us.”

Regulators scrambling to keep up

The plodding process of creating new regulatory frameworks typically lags well behind technological advancements and the gap has been well in evidence when it comes to the emerging artificial intelligence sector.

But 2025 saw plenty of efforts to play catch up, though the relative effectiveness of new state and federal guidelines has spurred plenty of debate thanks to the inherent tensions of consumer protections butting up against perceived roadblocks hindering AI advancements.

President Donald Trump, who made bolstering U.S. AI efforts a benchmark of his 2024 re-election campaign, wasted no time in taking action on the issue after taking the oath of office to begin his second term in early January.

On January 23, Trump issued an executive order, “Removing Barriers to American Leadership in Artificial Intelligence,” that included plans designating presidential advisors to coordinate with the heads of federal agencies and departments, among others, to develop an “action plan” to “sustain and enhance America’s global AI dominance” within 180 days, per a breakdown by the international law firm of Skadden, Arps, Slate, Meagher & Flom. The order arrived just days after Trump revoked Biden-era guidelines issued in 2023.

But the biggest federal move on AI regulation wouldn’t come until almost 11 months later.

On Dec. 11, Trump signed an executive order aiming to block state-level artificial intelligence regulations that he said will stifle the development and success of U.S. companies in the fast-advancing industry.

While a growing number of states, including Utah, enacted new laws in 2025 aiming to provide safeguards against potential AI-driven harms, federal lawmakers have so far failed to address widespread and bipartisan concerns over the risks of artificial intelligence tools.

During an Oval Office signing ceremony, Trump said he had spoken with leaders of the top U.S. tech businesses and was convinced that allowing state-level regulatory actions to proliferate will overcomplicate the playing field for domestic AI development efforts.

“We want to have one central source of approval,” Trump said. “Every time you make a change, and it could be a very reasonable change, you still won’t get it approved if you have to go to 50 states. So this centralizes it.”

The president also pointed to the global race for AI domination as a motivator for a universal, light-touch regulatory framework that is overseen by the federal government.

“I believe there will be only one country that benefits and it should be the United States,” Trump said. “And it will be if we are unified. China is unified because there is one vote and that’s President Xi.”

Where Utah stands on AI regulation efforts

Five new Utah artificial intelligence bills went into effect earlier this year with two of the bills, SB 226 and SB 332, amending 2024’s Artificial Intelligence Policy Act.

Passed in the 2025 session, HB 452 created new rules governing AI chatbots used in mental health treatment, while SB 271 expanded existing prohibitions against abuses of personal identity to include AI-generated content, like deepfakes. SB 180 created new requirements for state law enforcement agencies to keep a written AI policy and disclose the use of AI when used to generate documentation like police reports.

The Artificial Intelligence Policy Act adopted in last year’s legislative session established new, wide-ranging transparency requirements for businesses and licensed professionals to disclose when they use AI tools.

It’s not clear how Utah’s standing AI rules will be viewed in the Commerce Department’s assessment efforts but some critics believe the new executive order will face immediate legal challenges.

In a statement released following the announcement of Trump’s new AI executive order, Brad Carson, president of Americans for Responsible Innovation and a leader of the pro-AI regulation super PAC Public First, said Trump’s executive order is a “dud” politically and legally.

“This EO is going to hit a brick wall in the courts,” Carson said. “The executive order relies on a flimsy and overly broad interpretation of the Constitution’s interstate commerce clause cooked up by venture capitalists over the last six months. What’s more, this EO directly attacks the state-passed safeguards that we’ve seen vocal public support for over the past year, all without any replacement at the federal level.”

AI industry advocacy groups such as TechNet have previously argued that pausing state regulations would benefit smaller AI companies still getting on their feet and allow time for lawmakers to develop a country-wide regulatory framework that “balances innovation with accountability,” per a report from the Associated Press.

In a statement, TechNet president and CEO Linda Moore argues that AI “is transforming every sector of our economy” and that a state-based regulatory approach threatens “America’s ability to compete globally.”

“President Trump’s executive order on ‘Eliminating State Law Obstructing National AI Policy’ is an important step to ensuring there are clear and consistent rules that can keep Americans safe while allowing innovation to thrive,” Moore said. “States will still play a vital role, but a strong federal standard is essential to advancing AI innovation.”