In recent weeks, several companies have launched new shopping features on their AI platforms in time for the holiday season. Notable players include OpenAI, Google, Perplexity, and Microsoft, each offering various tools for consumers looking to make informed purchasing decisions. These tools aim to assist users by generating personalized buying guides, tracking prices, and even contacting local stores about product availability.
In testing these AI assistants, the author sought help in choosing a smartwatch compatible with the Nothing CMF Phone 1. They queried ChatGPT, Gemini, Perplexity, and Copilot over the same question: finding a suitable Android smartwatch. Each AI produced different results, revealing both strengths and weaknesses.
ChatGPT emerged as the most thorough assistant, offering detailed recommendations that included the Garmin Vivoactive 5 and providing side-by-side comparisons of features like design and battery life. However, it also suggested older models, such as the Fitbit Versa 4 and Google Pixel Watch 3. Despite being available, these older models lack some advanced features found in the newer Garmin Vivoactive 6, which raises questions about the relevance of these suggestions.
The author noted that while they did not specify a preference for the latest devices, the outdated recommendations could impact informed buying decisions. Google’s system returned no results for Garmin products after contacting local stores, highlighting a potential gap in its effectiveness.
Other AIs, such as Perplexity, suggested earlier models like the Samsung Galaxy Watch 4 from 2021, while also including less relevant products in search results. Copilot, on the other hand, successfully recommended the CMF Watch Pro 2, which is tailored for the Nothing CMF Phone 1, but failed to mention the newer CMF Watch Pro 3 commonly available in the market.
Overall, the AI shopping assistants demonstrated potential benefits but also faced limitations with offering current recommendations. Their performance raises questions about the reliance on AI for up-to-date information, suggesting users may still prefer traditional shopping guides or thorough product reviews.
Source: https://www.theverge.com/ai-artificial-intelligence/830450/ai-shopping-assistants-are-stuck-in-the-past