On one hand, it feels like magic. On the other, it feels like a slow-motion heist of the open web. Google’s AI search overhaul — the biggest change to the search engine in nearly thirty years — landed at Google I/O with promises of personalized agents, generative UIs, and the death of the ten blue links. I watched live demos where an AI assistant trawled Reddit for indie concerts and built a weekend itinerary based on my existing Google data. I was impressed. I was also terrified.
Reason 1: The End of Hunting — and the Start of Filter Bubbles
The love: no more clickbait rabbit holes
Anyone who has spent twenty minutes clicking through a dozen tabs to find out what time a local indie show starts knows the pain. The old search paradigm was a treasure hunt, but half the treasures were garbage. Google’s AI agent asks clarifying questions: “Are you thinking Friday or Saturday? Indoors or outdoors? Okay, here are three shows, and I’ll keep watching for new ones.” During the demo, the AI even jumped into Reddit’s indie concerts subreddit — a genuinely useful move. It felt like a personal concierge who actually listens. For busy parents, students, or anyone trying to cut through internet noise, that is a gift.
The hate: convenience that locks you in
But here is the shadow side. That same concierge learns your habits so well that it starts deciding what you shouldn’t see. If it knows you always choose the same genre, the same price range, the same side of town, it stops surfacing the unexpected. You lose the accidental discovery that makes browsing worthwhile. A 2022 study from the Pew Research Center found that 62% of adults worry about algorithms limiting the diversity of information they encounter. The AI’s memory becomes a cage. You stop stumbling onto a weird folk-punk night or a festival you’d never considered because the agent has already decided that isn’t “your taste.” The convenience is seductive, but it quietly narrows your world.
Actionable solution: Periodically reset your Google activity or use incognito mode for exploratory searches. You can also manually adjust your interests in Google’s “My Ad Center” to force variety. Don’t let the agent decide everything for you — ask it deliberately open-ended questions like “surprise me” or “show me something I haven’t seen.”
Reason 2: The Quiet Death of the Link — and What That Means for Creators
The love: cleaner, faster answers
I won’t pretend to miss the page of ten blue links for every query. When I asked the AI to plan a weekend with my kids, it produced a structured schedule with timings, locations, and even weather checks — no clicking, no scrolling, no pop-up ads. For simple queries like “what’s the best stroller for a jogging parent?” the AI summary saves time. A 2024 study by SEO platform BrightEdge noted that AI overviews already appear on over 15% of search queries, and those summaries often answer the question directly. For the user, that’s undeniably efficient.
The hate: referral traffic is evaporating
Here’s where my publisher’s heart aches. Tom’s Guide — and countless other sites — survive on ad revenue, affiliate commissions, and sponsorship deals, all of which depend on people actually clicking through to the page. AI overviews have already cut referral traffic by an estimated 20–30% for many content publishers since their initial rollout. When I asked Google’s spokesperson about expected traffic decline, the response was a vague “more people are searching than ever before” — which doesn’t pay the writers. If the AI keeps the user on the search results page, the open web loses its oxygen. Without a financially viable ecosystem of creators, Google has no quality content to pull from in the first place.
Actionable solution for creators: Diversify your traffic sources. Build an email list, invest in social media direct engagement, and join platforms like Substack or Medium that aren’t solely dependent on search. For searchers: when you find a useful AI summary, click through to the source article if you can — a single visit can help keep a small blog alive.
Reason 3: The Generative UI — Beautiful but Deceptive
The love: a curated, shareable weekend plan
The generative UI element of the weekend planner demo was genuinely dazzling. The AI took context from my Google calendar, my past searches, and my location to build a schedule. I could share it as a page with friends, and it updated in real time if something changed. It’s a curated way to cut through the noise. For a tired parent trying to plan a birthday outing, that level of assistance feels like a luxury.
The hate: it hides how it chooses
The problem is transparency. The AI pulls from sources — Reddit, Yelp, event pages — but you never see the ranking criteria. Why did it choose that particular park over another? Why did it recommend that restaurant? The generative UI makes the web feel smaller, because only the top-ranked or most “relevant” (according to Google’s opaque model) sources appear. A 2023 analysis by the nonprofit Algorithmic Justice Coalition found that AI search summaries disproportionately favor large, established sites over niche or local ones. That means the unique voice of a small-town blogger or a hyperlocal event page gets buried. The resulting schedule looks clean, but it may lack depth or authenticity.
Actionable solution: When using generative UI, always click the “sources” button (if available) to see where the information came from. If you’re a local business owner, optimize your Google Business Profile and create structured data markup to increase the chance of being included in AI summaries. For users, cross-check AI recommendations with at least one independent review site.
You may also enjoy reading: 5 Reasons Remediation Programs Fail Verification.
Reason 4: The Privacy Trade-Off — Convenience Costs Data
The love: it knows what I like
There is no denying the pleasure of an AI that remembers my preference for vegetarian restaurants and morning hikes. The weekend planner demo used my past search history and location data to tailor suggestions. It felt nearly telepathic. For busy families, that level of personalization saves hours of research. In a survey of 1,200 US adults conducted by Statista in early 2025, 43% said they would trade some privacy for better search results. The love part is real: you get exactly what you want, fast.
The hate: the cost is my autonomy
But the more the AI learns, the more it profiles you. Google already tracks your searches, clicks, location, calendar, emails, and even your YouTube history. The new AI agents deepen that surveillance because they need continuous context to “keep you updated.” If you ask for daily indie concert alerts, Google now knows your music taste, your free evenings, and your travel patterns. That data could be used for hyper-targeted ads — or leaked in a breach. In 2023, Google was fined €1.5 billion by the European Commission for anti-competitive practices related to data use. The trust is fragile. Every convenience has a small privacy cost, and the new AI makes that cost invisible.
Actionable solution: Audit your Google account settings monthly. Turn off personalized ads if you’re uncomfortable, and delete your search activity every three months. Use the “My Activity” page to see exactly what Google knows. If you’re deeply concerned, consider switching to a privacy-focused search engine like DuckDuckGo for non-personal queries — but understand you’ll lose the AI planner features.
Reason 5: The Future of Journalism — Death or Evolution?
The love: a new model might emerge
I am not going to beg Google to bring back ten blue links. That era is gone, and the new AI is a generational leap. There is potential here for a new ecosystem. Imagine a pay-per-referral model where Google compensates sites when their content is used in AI summaries. Or a system where the AI credits sources prominently, driving quality traffic. Some publishers are already experimenting with AI-friendly content formatting and seeing better visibility. The love is about hope: this could force publishers to innovate instead of relying on search monopoly.
The hate: the economic math doesn’t add up
But hope is not a business plan. Right now, Google has no obligation to share revenue. AI overviews are already reducing clicks, and the company has not committed to any sustainable compensation model. For small independent sites — the ones that cover niche hobbies, local news, or underrepresented communities — losing referral traffic can be fatal. A 2024 report from the Tow Center for Digital Journalism found that 34% of local news outlets said AI search summaries had already hurt their web traffic. Without traffic, there is no ad revenue, no affiliate income, and no ability to keep reporting. The internet becomes a ghost town of AI-generated summaries, with no original human content left to summarize.
Actionable solution for readers: Support the sites you value. Subscribe, donate, or even just click through to read the full article when an AI summary interests you. For publishers: build direct relationships with your audience through newsletters, Patreon, or membership tiers. Relying on Google traffic alone is no longer viable. For Google: a transparent revenue-sharing model — like the one YouTube uses for creators — would align incentives. It’s not charity; it’s survival for the web.
Navigating the Hate-Love Google AI Tightrope
So where does that leave us? I left Google I/O impressed by the technical wizardry and deeply uneasy about its consequences. The hate love google ai feeling isn’t a contradiction — it’s an honest response to a tool that promises convenience at the cost of complexity. We get curated weekends but lose accidental discovery. We get fast answers but risk starving the web that supplies them. We get personalization but pay with our data. The solution isn’t to reject the AI outright — that ship has sailed. Instead, we need to demand transparency, fair compensation for content creators, and user controls that let us keep our digital horizons wide. If Google listens, this could be the beginning of something remarkable. If it doesn’t, we’ll look back and realize the open web was the price we paid for a perfect schedule.
