AI Is Changing What People Think: 58% Say Chatbots Have Swayed Their Opinions
Here's an even more confusing one: **58% admit AI has influenced their opinions.**
We are living through a massive trust-use paradox. According to the new **2026 AI Consumer Insights Survey** from Shift Browser, artificial intelligence has quietly graduated from being a simple search tool to becoming an active persuader. We don't fully trust the technology, yet we're letting it shape our beliefs, our purchases, and our decisions.
Here's what the data tells us about our complicated relationship with the machines.
The Persuasion Engine
Traditional search engines gave us links; AI answer engines give us conclusions.
That shift is fundamental. When you search Google, you get a list of sources to evaluate. When you ask a chatbot, you get a confident, declarative answer. According to the survey of 1,448 nationally representative respondents, this confidence is working—even when it shouldn't.
* **58%** say AI-generated answers have influenced their opinions at least occasionally.
* **32%** use AI daily, meaning exposure to these persuasive nudges is constant.
* **53%** say AI improves their online experience, largely due to convenience.
The problem? Satisfaction doesn't equal accuracy. We love the streamlined experience of getting an answer in seconds without clicking through a dozen websites, even if that answer is incomplete or wrong.
The Trust-Use Paradox
If we know AI hallucinates, why do we listen to it?
The survey reveals a qualified confidence. While only 16% trust AI "a great deal," **60% trust it at least somewhat.** It's a dynamic similar to social media: we know the flaws, but we use it anyway because it's easy.
Accuracy is the second-biggest concern for users (36%), yet the convenience of an instant answer often outweighs the caution required to verify claims. We are prioritizing speed over truth.
The Black Box Problem
You can't verify what you don't understand.
A staggering **32% of respondents admitted they don't understand how AI systems generate answers.** This transparency gap is dangerous. Without insight into the process, users can't assess whether a response is well-sourced or speculative.
* **Privacy** is the top concern (48% worry about AI accessing personal data).
* **Transparency** ranks third (32%), directly linking to the inability to evaluate credibility.
* **44%** worry about AI taking actions without approval—including cognitive actions like steering opinions.
Traditional journalism and academic research come with attribution norms. AI-generated content often doesn't, leaving us to accept answers based on gut feeling rather than evidence.
Who Is Using AI?
The influence isn't spread evenly. Daily engagement is highest among **25- to 34-year-olds and working professionals.** These groups are integrating AI into workflows for research assistance (54%), summarization (34%), and automation (32%).
Conversely, adults 65 and older are the least likely to use AI (20% never engage with it). While this might insulate them from direct influence, it also leaves them less equipped to recognize AI-driven narratives when they encounter them secondhand through family or media.
The Demand for Control
People know the stakes are high. When a technology influences opinions at scale, accountability becomes urgent.
* **79%** favor some level of government regulation for AI answer engines.
* **35%** are calling for strong oversight.
* **57%** are concerned about the energy required to power AI systems, bringing environmental impact into the conversation alongside privacy.
Users want control, but current systems often hide it. **26% report difficulty managing or turning off AI features** once enabled. Tools that auto-enable features or bury opt-out mechanisms feed distrust and reduce users' ability to moderate AI's influence on their thinking.
The Bottom Line
AI answer engines are becoming information gatekeepers in ways search engines never were. A search engine returns options; an AI delivers conclusions.
The 58% who report AI influence may actually be underestimating the effect. Influence often operates below conscious awareness, shaping assumptions and framing issues without users realizing it. As Michael Foucher, VP of Product at Shift, noted: *"Consumers clearly see value in AI tools, yet they also want greater clarity and control over how those systems operate."*
As these tools become more sophisticated, the question remains: Will we build systems that inform without persuading, assist without steering, and answer without deciding? Or will we continue to let the machines make up our minds for us?