Every major technological shift seems to arrive with two competing promises.
The first is that life is about to become easier, faster and more connected. The second is that something distinctly human may be lost in the process.
Today, artificial intelligence sits at the centre of that debate. We are asking whether AI will weaken critical thinking, discourage original work, eliminate jobs and make people dependent on tools they do not fully understand.
But we have had a remarkably similar conversation before.
When Google and the wider internet became part of everyday life, many people worried that instant access to information would damage our memory, shorten our attention spans and reduce our willingness to think deeply. The technology was different, but the anxiety surrounding it sounds familiar.
Google also changed how we found almost everything.
Before search engines became part of everyday life, someone looking for a nearby plumber, restaurant or supplier might reach for the Yellow Pages. Google compressed that process into a search box. Instead of browsing a printed directory, people could describe what they needed, compare options and visit a business’s website within seconds.
Artificial intelligence may be beginning another version of that shift. The question is no longer only whether AI will change how we think. It is also whether it will change how we discover businesses, make decisions and take action.
Looking back at how we responded to Google may offer some perspective. Not because the technologies are identical, but because the comparison reveals something important: we often adapt more successfully than we expect, although we rarely emerge completely unchanged.
In 2008, writer Nicholas Carr captured a growing cultural anxiety with a provocative question: “Is Google Making Us Stupid?”
Carr argued that spending more time online was changing how he read and processed information. He found it harder to stay immersed in long pieces of writing and felt increasingly drawn toward scanning, searching and jumping between sources. His essay helped shape a larger debate about whether the internet was improving access to knowledge while weakening our ability to engage with it deeply.
The concerns extended well beyond reading.
Would students still learn facts if every answer could be searched? Would people lose the patience required to conduct proper research? Would easy access to knowledge make us more informed, or simply more confident in information we had barely examined?
Replace “Google” with “AI” and many of those questions still work.
Today, we wonder whether students will learn to write when an AI system can produce an essay in seconds. We question whether employees will develop expertise if a tool can summarize, analyze and draft on their behalf. We worry that people will accept a convenient answer without understanding how it was produced or whether it is correct.
The central fear is much the same: when technology performs part of our thinking for us, do we become less capable of thinking for ourselves?
It would be easy to conclude that the panic surrounding Google was exaggerated. Search engines became normal, society continued functioning and few people would return to a world where finding one piece of information required a trip to the library.
But that does not mean Google left our habits untouched.
A widely discussed study on the “Google effect” found that when people expected information to remain available, they were less likely to remember the information itself and more likely to remember where it could be found. The internet had begun functioning as a form of external memory.
That is not necessarily evidence that people became less intelligent. It suggests that we changed what we considered worth remembering.
Once information became instantly retrievable, knowing how to locate, compare and assess it became more important. We memorized fewer facts, but developed new skills around searching, judging sources and navigating vast amounts of information.
The result was neither complete decline nor uncomplicated progress. We gained extraordinary access to knowledge while also developing habits of distraction, scanning and dependence.
That is an important lesson for the AI era. Adaptation does not mean a technology has no negative consequences. It means people gradually develop new behaviours, expectations and skills around it.
Like Google, generative AI removes friction from tasks that once required more time and effort.
It can summarize a lengthy document, explain an unfamiliar concept, organize research, generate ideas and produce a first draft. Those abilities create opportunities, but they also raise questions about what happens when the effort disappears.
If we no longer struggle to find an answer, will we understand it as deeply?
If we do not create the first draft, will we still develop our own voice?
If a tool can propose a strategy, will we learn how to build one ourselves?
The unease is not limited to schools. AI is entering work that many people associate with intelligence, creativity and professional expertise.
Previous waves of automation were often discussed in relation to physical labour or repetitive administration. Generative AI can participate in writing, analysis, design, coding, planning and communication. These are activities many people once assumed technology could support, but not meaningfully perform.
The disruption therefore feels more personal. AI is challenging our assumptions about which parts of our work require a human mind.
The comparison with Google is useful, but it has limits.
Google primarily helped people locate information. The user still had to choose a result, interpret the source and assemble a conclusion.
Generative AI can collapse much of that process into one interaction.
Instead of directing us toward several sources, it can produce a polished answer. Instead of helping us research a report, it can draft the report. Instead of showing us examples, it can create something that appears complete.
That difference matters because a finished answer invites a different kind of trust than a list of search results.
When Google presents ten links, it is obvious that more work remains. When an AI system presents a clear and confident response, it can feel as though the work has already been completed.
The risk is not simply that AI may be wrong. It is that its output can be convincing enough to discourage further investigation.
AI may not remove critical thinking altogether, but it can shift it toward verifying information, identifying missing context and maintaining responsibility for the final result.
That distinction may help us understand the next stage of adaptation.
Google made it easier to find information. AI may make it possible to skip parts of the search process altogether.
Consider how someone chooses a product or service online. They might search for several options, open multiple websites, compare prices, read reviews and decide which business to contact.
An AI agent could combine those steps into a single request:
“Find a highly rated hotel near this conference that fits my budget.”
“Compare these software platforms and recommend the best one for my team.”
“Find a local contractor with commercial renovation experience and availability this month.”
The user may still visit a website before making a final decision, but that visit is no longer guaranteed to be the beginning of the journey. By the time the customer arrives, an AI system may have already identified the options, compared their differences and influenced which businesses receive consideration.
Early versions of this experience are already here. Search engines can generate summaries before presenting links. AI platforms can compare products, build itineraries, navigate websites and perform certain actions on a user’s behalf.
This does not mean that Google or business websites are about to disappear.
Websites will remain important places for demonstrating credibility, explaining complex services and building direct customer relationships. They will also continue to supply much of the information that search engines and AI systems rely on.
Their role within the customer journey may change, however.
For years, digital visibility has largely centred on one objective: get someone to click through to your website.
In an agent-driven environment, the objective may expand:
Make sure your business can be found, understood and accurately represented by the systems helping customers make decisions.
SEO is not becoming irrelevant. Websites still need useful content, clear technical structures, crawlable pages and accurate information.
But discoverability may begin to extend beyond conventional search rankings.
For ecommerce businesses, structured product feeds can help digital systems understand pricing, availability, product attributes and fulfilment options.
For service businesses, the equivalent may include clearly structured service information, current locations and hours, accessible booking systems and integrations that allow agents to check availability or initiate an action.
This does not mean every business needs to build a custom API or abandon its current SEO strategy. It means businesses should think beyond how their information looks to a human visitor and consider whether digital systems can interpret it accurately.
A beautifully designed website is less useful if its services are difficult to understand, its important information is buried or its systems cannot communicate with the platforms customers rely on.
The goal is not to design for machines instead of people. It is to create digital experiences that work well for both.
Google changed the valuable skill from simply knowing information to knowing how to find and assess it.
AI may be creating another shift.
As tools become more capable of producing content and recommendations, human value may move further toward defining the problem, providing context, evaluating the response and deciding what should happen next.
Writing a first draft may take less time, but recognizing when that draft is generic or inaccurate still requires judgment.
Producing ideas may become easier, but choosing one that fits the audience and situation still requires understanding.
Summarizing information may become nearly instant, but deciding which information matters still requires expertise.
The danger is not that AI performs the earlier steps. The danger is that we mistake completing an earlier step for completing the entire task.
A useful AI workflow should not remove the human from the process. It should change where human effort is concentrated.
The most valuable skills may include asking precise questions, recognizing weak reasoning, checking claims, identifying missing context and taking accountability for the final decision.
These skills may be less visible than producing a page of writing from scratch, but they are not less valuable. They are often the difference between output that looks impressive and work that is genuinely useful.
The history of Google gives us a reason to be optimistic, but it does not give us permission to be passive.
People eventually developed search literacy because they had to. Schools began teaching students how to evaluate online sources. Businesses created standards for protecting information. Users became more aware of misleading websites and unreliable content.
The same process now needs to happen with AI.
Organizations need clearer expectations around where AI can be used, what information can be entered into a system and when human review is required. Teams need to understand both what these tools do well and where they are likely to fail.
Individuals also need foundational skills before delegating work. Someone who has never learned how to research, write or solve a particular problem may struggle to recognize when AI has produced a poor result.
AI literacy should involve more than learning how to write better prompts. It should include knowing when not to use AI, when to question it and when the consequences of an error require greater human oversight.
For businesses, adaptation does not mean chasing every new tool or integration. It means understanding how customer behaviour is changing and deciding which changes genuinely matter.
Technology often inspires extreme predictions because its long-term effects are difficult to imagine while its most disruptive qualities are immediately visible.
Google was going to destroy memory. Smartphones were going to end meaningful conversation. Social media was going to democratize information. Each prediction contained part of the truth, but none captured the entire future.
People adapted. New behaviours, industries and skills emerged. Some capabilities became less important, while others grew in value. At the same time, some of the original concerns proved justified.
AI will likely follow a similarly complicated path.
It may reduce the effort required for certain tasks while raising expectations for speed and productivity. It may make expertise more accessible while making false confidence easier to acquire. It may allow people to focus on more valuable work, but only when they understand the work well enough to decide what should be delegated.
It may also reshape how people discover products, evaluate services and interact with businesses.
The Yellow Pages did not disappear because people stopped looking for local businesses. The need remained, but the interface changed.
Google became the new starting point. Businesses adapted by creating websites, learning how search engines worked and competing for visibility.
AI may now be changing the starting point again.
The next stage of adaptation may not only involve learning how to use AI within our work. It may also require businesses to understand how AI systems find information, compare options and act on behalf of potential customers.
Google taught us how to find information in a world where knowledge became instantly accessible. It also taught businesses how to become discoverable in a world organized by search.
AI is asking harder questions. When a machine can produce an answer, which parts of the process should still require human thought? When an agent can help a customer make a decision, how does a business make sure it is found, understood and accurately represented?
How we answer those questions will determine whether AI becomes a substitute for our abilities or a tool that helps us apply them more effectively.