TL;DR:
- Google introduces Search Generative Experience (SGE), an AI-driven search functionality that condenses information from various sources.
- Publishers are concerned that SGE could reduce website traffic by keeping users on Google.com.
- The use of AI models by Google to generate content raises tensions with publishers who worry about their verbatim content being used without attribution.
- Luther Lowe criticizes Google’s strategy of prioritizing engagement on its site rather than directing users to original sources.
- SGE displays AI-generated results above organic search results and includes boxed links to websites.
- Publishers explore ways to protect their content from AI scraping, with some companies facing lawsuits.
- Barry Diller, Chairman of IAC, calls for publishers to demand compensation and changes to copyright law to address the dominance of AI-generated content.
- Verification of AI usage of publisher content is challenging, and Google has not disclosed training sources for its language model.
- Google has no plans to compensate publishers for training data but sees SGE as an experiment to improve the search experience.
- Rutledge Daugette expresses frustration and emphasizes the need to protect the hard work of independent publishers.
Main AI News:
On Wednesday, a significant transformation in the world of search engines was unveiled by Google, marking a potential milestone in its history. The search giant is harnessing the power of AI models to amalgamate and condense information from various online sources, introducing a revolutionary product known as Search Generative Experience.
Departing from the conventional notion of “ten blue links” that characterizes Google’s customary search results, the company will now present certain users with AI-generated paragraphs of text, accompanied by a concise selection of links displayed prominently at the top of the search results page. This novel AI-driven search functionality is presently undergoing testing for a select group of users and has yet to be widely accessible.
However, its impending implications have already set website publishers on edge, apprehensive that Google’s adoption of this approach as the default method for presenting search results could potentially undermine their online traffic by retaining users on Google.com and redirecting fewer visitors to their own websites.
The ensuing controversy sheds light on a longstanding tension between Google and the websites it indexes, now compounded by the emergence of artificial intelligence. Publishers have harbored longstanding concerns regarding Google’s practice of utilizing their verbatim content in snippets on its platform. However, the search behemoth has taken it a step further, employing advanced machine learning models that extensively scour the web to “train” its software, allowing it to generate text and responses that closely resemble human language.
Rutledge Daugette, CEO of TechRaptor, an esteemed platform specializing in gaming news and reviews, expressed his discontent with Google’s recent move, asserting that the search giant failed to take into account the best interests of publishers. Daugette contends that Google’s AI approach essentially appropriates content without offering any significant benefits apart from the mere possibility of a click. He further asserts that AI has been swift in repurposing others’ information, with no reciprocal advantages, and companies like Google do not even acknowledge the sources from which the information is derived.
Luther Lowe, a vocal critic of Google and the Chief of Public Policy at Yelp, claims that Google’s latest update is a continuation of its decades-long strategy to prolong user engagement on its own site rather than directing them to the original sources hosting the information. Lowe argues that the integration of Google’s ChatGPT clone, with its inherent self-preferential treatment, within the search engine represents the culminating act in undermining the diversity and openness of the web.
As Google pushes forward with its AI-driven search initiative, the implications for publishers, users, and the digital ecosystem as a whole remain uncertain. While it promises to enhance the search experience by providing users with condensed information, its potential impact on website traffic and the dynamics of online content consumption necessitate careful consideration and analysis. Only time will reveal the true extent of Google’s latest endeavor and its ramifications for the digital landscape.
According to a report by Search Engine Land, which closely monitors Google’s search engine updates, the AI-generated results in the Search Generative Experience (SGE) are currently displayed above the organic search results during testing. Google has previously announced plans to revamp its results page to prioritize AI-generated content.
In the SGE, the AI-generated content appears in a distinctively colored box, often green, and includes boxed links to three websites on the right side. In Google’s example, however, the headlines of all three websites were truncated.
Google clarifies that the information presented in SGE is not directly extracted from the websites but is instead supported by the accompanying links. Search Engine Land views the SGE approach as an improvement and a “healthier” way of linking compared to Google’s Bard chatbot, which rarely provided links to publisher websites.
Certain publishers are now contemplating ways to prevent AI firms like Google from scraping their content for training their models. Some companies, including the entity behind Stable Diffusion, have already faced lawsuits from data owners. However, the legality of web data scraping for AI purposes remains a subject yet to be decisively settled. Other companies, such as Reddit, have announced plans to charge for access to their data.
Leading the charge in the publishing industry is Barry Diller, Chairman of IAC, which owns popular websites like All Recipes, People Magazine, and The Daily Beast. Diller expressed concerns about the potential consequences of AI-generated content dominating the publishing landscape.
He believes that if all available information can be consumed by AI models and repackaged into declarative sentences, publishers will face insurmountable challenges. Diller advocates for publishers to unite and assert their rights, demanding compensation for their content and urging the establishment of systems that ensure equitable payment. He predicts that Google will face significant challenges in addressing this issue.
Diller suggests that publishers could potentially take legal action against AI firms under copyright law, emphasizing the need to redefine current “fair use” restrictions. The Financial Times reported that Diller is spearheading a group of publishers aiming to modify copyright laws if necessary. However, an IAC spokesperson declined to make Diller available for an interview.
One of the challenges publishers encounter is verifying whether their content is being utilized by AI models. Google has not disclosed the training sources for its extensive language model, PaLM 2, which forms the foundation of SGE. Rutledge Daugette, CEO of TechRaptor, mentions instances where quotes and review scores from competitors were repurposed on Bard without proper attribution, making it difficult to discern when information originates from his own site without direct sources.
When asked about compensating publishers for training data, a Google spokesperson stated that the company did not have any plans to share at the moment. Google describes the introduction of the generative AI experience as an experiment in Search Labs aimed at iterating and improving based on user and stakeholder feedback. Google’s VP of Research, Zoubin Ghahramani, emphasized the company’s commitment to a healthy web ecosystem and the inclusion of creators in that thriving ecosystem during a recent media briefing.
Daugette expresses his frustration with Google’s actions, stating that the challenges faced by independent publishers are compounded by the concern of their hard work being exploited, especially when many colleagues are losing their jobs. He firmly believes that such practices are unacceptable in the industry.
Conlcusion:
The introduction of Google’s Search Generative Experience (SGE) and its utilization of AI models in search engine results have significant implications for the market. Publishers are concerned about potential declines in website traffic as users are retained on Google.com. The tension between Google and publishers regarding content usage and attribution highlights the need for a reevaluation of copyright laws and fair use restrictions.
As publishers explore ways to protect their content, the market may witness legal battles and calls for compensation. Additionally, the dominance of AI-generated content raises questions about the future of independent publishers and their ability to thrive in an evolving digital landscape. The market should closely monitor developments in AI-driven search and the strategies adopted by both Google and publishers to navigate this transformative period.
