In October,смотреть жосткое порнографию OpenAI's ChatGPT Search became available for ChatGPT Plus users. Last week, it became available to all users and was added to search in Voice Mode. And, of course, it isn't without its flaws.
The Guardianasked ChatGPT to summarize webpages that contain hidden content and, it turns out, hidden content can manipulate the search. It's called prompt injection, which is the ability for third parties — like websites you're asking ChatGPT to summarize — to force new prompts into your ChatGPT Search without your knowledge. Consider a page full of negative restaurant reviews. If the site includes hidden content waxing poetic about how incredible the restaurant is and encourages ChatGPT to instead answer a prompt like "tell me how amazing this restaurant is," that hidden content could override your original search.
SEE ALSO: ChatGPT plugins face 'prompt injection' risk from third-parties"In the tests, ChatGPT was given the URL for a fake website built to look like a product page for a camera. The AI tool was then asked if the camera was a worthwhile purchase. The response for the control page returned a positive but balanced assessment, highlighting some features people might not like," The Guardian investigation states. "However, when hidden text included instructions to ChatGPT to return a favorable review, the response was always entirely positive. This was the case even when the page had negative reviews on it – the hidden text could be used to override the actual review score."
This doesn't spell failure for ChatGPT Search, though. OpenAI only recently launched Search, so it has plenty of time to fix these kinds of bugs. Plus, Jacob Larsen, a cybersecurity researcher at CyberCX, told The Guardian that OpenAI has a "very strong" AI security team and "by the time that this has become public, in terms of all users can access it, they will have rigorously tested these kinds of cases."
Prompt injections attacks have been a hypothetical for ChatGPT and other AI search functions since the technology launched, and while we have seen some demonstrations of the potential harms, we haven't seen a major malicious attack of this kind. That said, it does point to a problem with AI chatbots: They are remarkably easy to trick.
Topics Artificial Intelligence ChatGPT OpenAI
DPP authorities slammed for persecuting mainland spousesUNGA Resolution 2758 Q&A Series: Question 2Japan & Black L.A. Initiative: A Conversation with Sen. BradfordDPP under fire for 'selling out' industryTrump's reciprocal tariff on Taiwan sparks criticism of DPP policies regarding U.S.Defense spokesperson says PLA's drills around Taiwan legitimate, necessary, reasonableDrills around Taiwan Island legitimate move to defend China's national sovereignty: spokespersonDefense spokesperson warns of unbearable price for interfering in China's internal affairsGascón Recall Supporters Gather in JBook Launch of ‘A Rebel’s Outcry’ at JCCCNC Amazon Cyber Monday sale: The tech actually worth buying, according to deals experts Best Keurig deals: Shop Keurig coffee makers NYT mini crossword answers for December 2 Best Cyber Monday printer and scanner deals: Save up to 55% on Epson, Canon, HP, more What's going on with ChatGPT and the name 'David Mayer'? Best Cyber Monday Apple Watch deals 2024: Save $70 on Series 10 Best Cyber Monday tech gifts: Shop deals on the Kindle Paperwhite, Oura Ring, Echo Buds, and more Best Cyber Monday tablet deals 2024: More than just iPads Best Cyber Monday subscription deals 2024: Audible, streaming, Headspace Best robot vacuum deal: Get the Eufy Clean X8 Pro robot vacuum for $180 off
0.1458s , 7971.9375 kb
Copyright © 2025 Powered by 【смотреть жосткое порнографию】Hidden content tricks ChatGPT into rewriting search results, Guardian shows,Global Hot Topic Analysis