Alphabet founders saw their on-paper net worth jump on news of Google's introduction of a Search Generative Experience (SGE) powered by artificial intelligence (AI), as reported by Bloomberg earlier today. But the move, which follows a similar one by Microsoft's Bing earlier this year, could trigger great opportunity or a significant industry conflict as publishers worry about the potential loss of website traffic and financial implications, according to a CNBC report. This AI-based update may replace the traditional list of search results, or "ten blue links", with AI-generated summaries and a few links at the top of the results page.
Although SGE is currently being tested with a select group of users and not widely available, publishers are expressing concerns that should SGE become Google's default search experience, it may diminish the number of users visiting their sites and instead retain them on Google.com, as we previously reported. Furthermore, publishers are questioning whether they can prevent AI companies like Google from scraping their content for training purposes, with some considering legal options for compensation, according to a Wall Street Journal article.
Artificial intelligence systems such as Google's larger language models that form the foundation of SGE, ChatGPT, and other search engine improvements by tech giants are trained by scraping vast amounts of data from websites, generating concerns amongst publishers that their copyrighted material is being repurposed by these AI models
With our new generative AI experience in Search, you’ll get even more from a single search. You’ll be able to quickly make sense of information with an AI-powered snapshot, pointers to explore more and natural ways to ask. #GoogleIO pic.twitter.com/jgzz97DzEv— Google (@Google) May 10, 2023
As reported by Search Engine Journal, SGE offers a search results page displaying AI-generated text in a differently colored box, with links to source websites on one side. While Google maintains that the AI-generated content is corroborated by these links, and not directly taken from websites, publishers worry that user traffic will decline as fewer individuals click on these links, opting instead for the AI-generated summaries.
One concern in the publishing world is the possibility that AI-generated search results from Google might not always be accurate, as seen in a Washington Post article, which highlighted an example in which Google's AI provided an incorrect chocolate chip cookie recipe. Large-scale adoption of AI-generated search results may lead to further questioning of the accuracy and bias in AI-generated content.
Industry heavyweights like Barry Diller, Chairman of IAC, believe that publishers can use copyright law to prevent AI companies from scraping their content without compensation. He maintains that current "fair use" restrictions should be redefined, and has spearheaded a group of publishers that are looking to change copyright laws if necessary, as reported by the Financial Times.
Some publishers, such as Reddit, are exploring the idea of charging for access to their data, as AI technologies blur the lines between fair use and copyright infringement. Future legislation and legal cases will likely play a significant role in determining the extent to which AI companies can use web-scraping techniques to train their models, dramatically impacting the future of online search.
Amid this ongoing industry debate, many other publishers say they are embracing the new change, as they believe the A.I. search might have beneficial outcomes, even for them. One industry veteran the Hoodline spoke to believes that there is a chance it will allow users to find the best article more effectively, thereby bringing greater traffic to their news stories and more revenue to their bottom line. Others suspect that the future could hold deeper partnerships between publishers and the search or social media platforms through which users find their news.
Regardless of how AI-driven search experiences progress, the impact on online publishers and the digital economy at large will be profound, with decision-making processes and legal considerations likely to have lasting effects on the future of online search as we know it today, all of it culminating in what seems to be an inevitable showdown, and the consequences will surely reverberate throughout the industry, and very-well society, in the years to come.