This may in fact be a huge opportunity for dedicated publishers.

I’m not referring to publishers contributing to Google and OpenAI, nor to publishers creating their own AI summaries in an evident attempt to become AI technologists. (Who believes they can outperform Google and OpenAI in their own domain?)

I’m discussing the unique strengths of publishers, the opposite perspective.

Recently, there’s been much debate over Google AI’s erroneous “AI overview” results, such as pizza topped with glue. For many commentators, this signaled Google’s premature product launch, a hasty attempt to outpace AI competitors, with the assumption that it will eventually improve significantly, if not achieve perfection.

However, I wonder: Can Google truly accomplish this?

Google excels at delivering relevant content to any query. Yet, it remains the human’s responsibility to evaluate those results and discern which are trustworthy. Thankfully, we’ve learned to do this, though not everyone is equally adept.

Another consideration is the diversity of opinions on any given topic. Critical thinkers seek various perspectives to form their own views. Unfortunately, some of this diversity is arguably already lost within the “opinion bubbles” prevalent on social media, and we’re all feeling the impact.

Trusting them to know the truth

Trusting a company like Google with AI-generated summaries means entrusting them to discern truth, to sift through myriad results—including the inevitable inaccuracies. At a minimum, it implies relying on Google to encapsulate diverse opinions and present nuanced views on a topic without omitting crucial information.

With 45 years in IT, starting with punched paper tape, I’m optimistic about technological possibilities. Satirical content, like that from The Onion, could be flagged or recognized, potentially resolving issues like the eating rocks and the pizza scenario. It’s also worth remembering that humans can fall for these errors too.

But the fundamental question I pose is:

Do we truly desire this?

Should we allow a corporation like Google to become even more of an arbiter of truth than it already is? Should we let a machine, developed and operated by a faceless team of technologists, determine what is true or false, right or wrong, or what transpired at a particular event, or even if it occurred at all?

It’s at this juncture that I begin to appreciate the human journalist, the discernment of a competent editorial team, and the credibility of a publishing brand that stands to lose a great deal if it were to disseminate falsehoods or be implicated in fraud.

For instance, when Sam Altman urgently needed to demonstrate that he had not misappropriated Scarlett Johansson’s voice, he did not resort to GPT-40. He approached the Washington Post, requesting that they conduct an investigation and report their findings.

This instance could serve as a prime illustration of why serious publishers and seasoned, knowledgeable journalists ought to remain the arbiters of, if not truth, then at least the reliability of information available to the public.

Update: From Ars Technica (“Google’s AI Overview is flawed by design, and a new company blog post hints at why“):

Here we see the fundamental flaw of the system:

“AI Overviews are built to only show information that is backed up by top web results.” [quote from Google Blog].

The design is based on the false assumption that Google’s page-ranking algorithm favors accurate results and not SEO-gamed garbage. [emphasis mine]. Google Search has been broken for some time, and now the company is relying on those gamed and spam-filled results to feed its new AI model.

So it seems to work very similar to the Bing AI Overviews. It takes a result list from the well-known search engine and writes a summary about it. This is technically different from GPT, yet both suffer from the same fundamental problem:

They simply don’t know the truth.

We believe that, to compete with tech giants, publishers need the best technology for themselves. But developing in-house to compete with Google is not realistic, not even for large publishing houses. That is why, for over 30 years, COMYAN has been committed to developing technologies that enable publishers to transition into the digital era successfully. We are not competing for users, readers or advertisers. Our mission is to supply the best tools and the best know-how to make publishers successful as digital publishers in the future.

To learn more about our products, services, and consulting, please reach out to us now: