This week, Avram Piltch discusses his concerns about the use of AI to generate and present information as an arbiter of truth. He specifically talks about Google's AI overview, previously known as Google SGE, which has recently been rolled out to everyone in the United States. He expresses his strong feelings about the potential for AI-generated information to be misleading, especially when presented as authoritative facts without acknowledging differing viewpoints or ongoing legal disputes.
He gives an example of a search query about whether training data is considered fair use under US copyright law. The AI overview provided by Google confidently states that training AI models using publicly available internet materials is generally considered fair use. However, Avram points out that this statement is presented as a fact, without mentioning the ongoing lawsuits against OpenAI and the debate surrounding the issue.
He highlights the danger of people trusting AI-generated information blindly, assuming that Google's answers are always correct. He emphasizes the importance of presenting information in a balanced and objective manner, acknowledging different perspectives and ongoing legal disputes.
Avram has created a browser extension that allows users to skip the AI-generated answers and get right to the results. The extension, called Bye Bye, Google AI uses CSS manipulation in order to suppress the AI areas of the search results. This approach allows the extension to work without the user having to do anything special.
One of the key benefits of customizing search results with CSS is the ability to personalize the browsing experience and filter out irrelevant or distracting content. Avram discusses how he was inspired to create his own browser extension after seeing another extension that allowed users to hide elements on Google search results pages. By developing his own extension, He was able to provide users with more options for customizing their search experience and tailor the results to their liking.
However, Avram also acknowledges the limitations of using CSS to customize search results. He notes that Google can update its algorithms at any time, potentially breaking the functionality of his extension and causing unexpected issues, such as making certain elements disappear from the page. Despite these challenges, he remains committed to refining his extension and addressing any issues that may arise, demonstrating his dedication to providing users with a valuable tool for customizing their search experience.
Avram discusses the issue of AI-generated search results and offers a solution for those who prefer to avoid them. He introduces the concept of the "web tab," which presents search results in a more traditional format without AI-generated content such as featured snippets, videos, and discussions. He acknowledges that while AI technology can be helpful, it is not always accurate or reliable, leading some users to seek alternatives.
He also provides a workaround for accessing the web tab directly without the need for his extension. By creating a custom search engine in the browser settings with the parameter "UDM=14," users can bypass AI-generated search results and go straight to the web tab. This allows users to access a more traditional search experience without the influence of AI-generated content.
However, Avram notes that this solution may not be as easily accessible on mobile devices, as popular mobile browsers like Chrome and Safari do not offer the same customization options for search engines. While Firefox does provide a feature for adding custom search engines, users of other mobile browsers may have limited options for avoiding AI-generated content.
AI-generated search results can be misleading. Avram discusses his experiences with AI-generated search results and highlights the potential for misinformation and inaccuracies to be presented as authoritative truth. He notes that AI-generated responses may not always provide accurate information and can sometimes offer questionable advice or opinions.
One example Avram mentions is receiving conflicting advice on how to prevent or cure kidney stones, with some responses suggesting drinking urine as a remedy. This highlights the potential dangers of relying on AI-generated information without proper verification or critical thinking. Additionally, he discusses instances where AI-generated responses offer incorrect or potentially harmful advice, such as recommending opening a camera to remove stuck film.
Furthermore, Avram points out that AI-generated responses may not always consider differing viewpoints or present information in a balanced manner. For example, in cases where there is controversy or debate, AI may present a single opinion as fact without acknowledging opposing perspectives. This can lead to a skewed presentation of information and misinform users who rely on AI-generated search results.
Overall, Avram's discussion underscores the importance of being critical consumers of information and questioning the reliability of AI-generated search results. While AI technology has the potential to provide valuable insights and assistance, it is crucial for users to verify information, consider multiple sources, and exercise caution when relying on AI-generated responses. By being aware of the limitations and potential pitfalls of AI-generated search results, users can better navigate the digital landscape and avoid falling victim to misinformation and inaccuracies.
For more information on AI in Google Search, check out Avram's full article.
Scott is a developer who has worked on projects of varying sizes, including all of the PLUGHITZ Corporation properties. He is also known in the gaming world for his time supporting the rhythm game community, through DDRLover and hosting tournaments throughout the Tampa Bay Area. Currently, when he is not working on software projects or hosting F5 Live: Refreshing Technology, Scott can often be found returning to his high school days working with the Foundation for Inspiration and Recognition of Science and Technology (FIRST), mentoring teams and helping with ROBOTICON Tampa Bay. He has also helped found a student software learning group, the ASCII Warriors, currently housed at AMRoC Fab Lab.
Avram's been in love with PCs since he played original Castle Wolfenstein on an Apple II+. Before joining Tom's Hardware, for 10 years, he served as Online Editorial Director for sister sites Tom's Guide and Laptop Mag, where he programmed the CMS and many of the benchmarks. When he's not editing, writing or stumbling around trade show halls, you'll find him building Arduino robots with his son and watching every single superhero show on the CW.