When you need to find something online, do you say you have to “search” for something or “Google it”?

I worked in the search industry before Google existed (circa 1998)… and I say, “Google it.”
When something becomes that big and the business landscape changes, we look for the errors as a sign of something else (which isn’t always the smartest strategy).
Google’s latest search feature, AI Overviews, was meant to provide their AI chatbot’s answers at the top of the organic search results page.
And while it has been impressive to see the integration, there are (obviously) errors and “hallucinations” that have led to a slew of bizarre and inaccurate answers (and guess which story gets more headlines?).
I’d argue that the concept behind AI Overviews is solid: Leverage generative AI to deliver more direct and conversational search results.

I also knew it would get wonky.

From suggesting glue on pizza to keep the cheese in place to recommending eating rocks, the AI’s brainzaps have been both humorous and alarming (to some).
I get it.
It’s not just about a few goofy suggestions… it’s about the credibility and reliability of a tool that millions depend on to be accurate.

So, what’s going on?

The crux of the issue lies in AI’s inability to distinguish between serious content and jokes, comments, sarcasm, etc. that us users love to clapback on to any form of online content (hello, Troll).
Like sourcing information from platforms like Reddit, where sarcasm and humor reign supreme.

What is clear and correct?

While AI is fast and powerful, it’s not yet nuanced enough to parse context effectively (sadly, I’d argue that many people that I interact with on a daily basis are equally challenged by sarcasm and dry humor).
Still, let’s give credit where credit is due…
Google has been quick to address these issues.
They’re working to refine their algorithms and improve the AI’s understanding.
The real-time feedback loop, while messy, is essential for the AI’s evolution (and something other brands – beyond tech – could learn from).

Google’s push for AI Overviews isn’t happening in a vacuum.

The rise of competitors like OpenAI and Microsoft may be adding pressure.
Microsoft CEO, Satya Nadella’s famously said: “We made them dance.”
Google, despite its vast experience and resources in AI (remember, they own DeepMind and provided the paper on Transformers that led to OpenAI’s developments), seems to be playing catch-up in the public square.
Google has been inching towards this moment for years, integrating features like images, video, and even ads into search results.

But this leap to conversational AI feels different.

It’s about creating an intuitive, human-like interaction while we search for content that moves us away from multiple different sources and to one answer that Google provides.

What’s next for AI in search?

There’s a delicate balance between making search results more interactive and maintaining the trustworthiness of those results (a struggle that Google has been battling since the term “Search Engine Optimization” was christened.
The idea of a universal AI assistant like Google’s Project Astra, which can seamlessly integrate into our daily lives – talk to us and assist us – must be intoxicating for anyone tinkering with the next version of generative AI.

So what do we have to show for it today?

It’s a reminder that even in a world driven by algorithms and machine learning, the human touch – and a bit of skepticism – remains indispensable.
I’m optimistic about Google and the future of search… and not surprised by anything that’s been happening at these early stages of this journey.

This is what Elias Makos and I discussed on CJAD 800 AM. Listen in right here.

Before you go… ThinkersOne is a new way for organizations to buy bite-sized and personalized thought leadership video content (live and recorded) from the best Thinkers in the world. If you’re looking to add excitement and big smarts to your meetings, corporate events, company off-sites, “lunch & learns” and beyond, check it out.