A Billion Users, and Most of Them Are Nervous About It

Key points

• ChatGPT reached 1 billion monthly users in May 2026, according to market analysis firm Sensor Tower.
• More US adults now expect AI to have a negative personal impact than a positive one, according to the Pew Research Center.
• Google DeepMind's Gemini logged 950 million monthly users in July 2026.
• All 50 US states have passed or proposed AI laws, producing more than 2,100 bills, a tenfold increase in three years.
• A Stanford University report found more than half of people worldwide say AI products make them nervous.

ChatGPT is now used by one in every eight people on Earth. Set that beside a Pew Research Center finding: a majority of American adults believe AI will hurt them, not help them. Both things are true at once.

Pew also found that half of US adults now use a chatbot (a text-based AI assistant) more than double the share who said so in 2023. One in four say they use one every day. Across the 38 OECD countries, representing most of the world's wealthiest democracies, more than a third of adults say they used a generative AI tool in the past three months. Our coverage of a Pew poll of more than 42,000 people across 37 countries, published 19 September, found the same worry running even among people who have never touched an AI tool.

Why do people keep using something they distrust?

The frustration is mostly directed at companies, not the tools. The objection is to the relentless corporate push to insert AI into every product while simultaneously warning that the technology will reshape the global economy. The tools themselves often still feel useful.

A Gallup poll from May found 71 percent of US adults would oppose a new AI data centre being built near them. That's a higher opposition rate than for a new nuclear power plant, at 53 percent. The pessimism runs deepest among younger adults, the same group that adopted smartphones fastest a decade ago.

There's a geographic pattern too. Heavy adoption and heavy skepticism tend to arrive together: the Global North skews pessimistic, where uptake is highest, while people in the Global South, where adoption is lower, remain more optimistic.

Is this just the social-media story again?

Partly, but the exit routes are different. When Facebook and Twitter drew widespread criticism, users faced a stark choice: stay or lose every connection and piece of content they'd built over years. Most stayed.

AI doesn't yet have that lock-in. Several strong open-source alternatives to ChatGPT and Gemini already exist. Open-source means the underlying code is publicly available, so anyone can run or modify it without paying OpenAI or Google. Market pressure on the big players is still real, for now.

Regulation is also moving faster. Every US state has either passed or proposed an AI law. The resulting patchwork of more than 2,100 bills represents a tenfold increase in three years, giving consumers tools they never had during social media's formative years. That's worth keeping in mind practically: you're not obligated to use the first AI product a company puts in front of you. Switching costs are still low, and that window may not stay open forever.

The billion-user milestone is real. So is the unease. The companies that figure out how to earn trust rather than simply manufacture scale will be the ones worth watching.

Common questions

Does using AI mean you support it?

Not necessarily. Surveys suggest many people use AI tools for particular tasks while still holding reservations about how the technology is being developed and deployed at scale.

What do the new state AI laws actually do for ordinary people?

Most bills focus on transparency and accountability: requiring companies to disclose when AI makes a decision that affects you, such as a loan application or a job screening, and giving you some right to appeal. Effects and enforcement dates vary by state.