Many commentators have highlighted the competitive risk large language models (LLMs) like ChatGPT pose to Google's search platform. Consumers, they rightly point out, would prefer to be spoon-fed 'the answer(tm)' rather than a bunch of DIY links to pages that may contain an answer.

LLM developers, they say, have left Google flatfooted and it is struggling to respond.

But they are missing the larger point.

Google pioneered LLMs and has sophisticated technology of its own - it just cannot work out how to monetise it.

>>
Google has spent several years working on chat bots and, like other big tech companies, has aggressively pursued artificial intelligence technology. Google has already built a chat bot that could rival ChatGPT. In fact, the technology at the heart of OpenAI’s chat bot was developed by researchers at Google.
...
Google may be reluctant to deploy this new tech as a replacement for online search, however, because it is not suited to delivering digital ads, which accounted for more than 80 percent of the company’s revenue last year.
>>

Google monopolised the ad word market because it recognised early on that being the (free) search tool of choice provided it with incredible, real-time insights into consumer intent - at a highly granular, individual level, against which it could sell advertisements.

The struggle for Google is how it can ensure it remains firmly inserted between consumers and vendors (or, at the very least, reliably assert its role in the vendors' attribution model) if it provides 'the answer(tm)' rather than (merely) search results.

There is also the prickly issue of whether LLM-generated responses (answers) are akin to editorial, potentially exposing Google (as publisher) to liability for inaccuracies/errors and aggravating current/potential advertisers when recommending Product A over Product B.

#artificialintelligence #technology #digital #tech #chatgpt #google #advertisingandmarketing #llms