The BBC tested four artificial intelligence models to see whether they could help shoppers get a better price on a new television ahead of Black Friday. The short practical experiment set out to assess how well widely available AI systems can identify discounts, compare offers and steer buyers toward a cheaper purchase.
Testers asked each model to locate current deals and recommend an option, then compared the AI responses with live prices and conventional shopping tools. The trial focused on practical shopping tasks users often want help with: finding lower prices, spotting limited-time offers and weighing value between different screen sizes and features. The exercise was designed to mirror how a typical consumer might use a chatbot or virtual assistant while shopping.
The trial highlighted both the potential and the limits of current AI tools in a fast-moving retail context. Artificial intelligence can quickly surface suggestions, explain technical specifications in plain language and flag areas to check, such as warranty or return policies. However, models that do not have up-to-the-minute access to retailer databases or the web can miss the latest price changes, and all can produce confident-sounding answers that need verification.
For shoppers, the takeaway is to treat AI as a supplementary research aid rather than a definitive arbiter. Use it to narrow options and clarify product differences, but confirm final prices, coupon validity and stock levels directly with retailers or price-tracking services before buying. As Black Friday approaches, combining AI-assisted comparison with established tools — retailer websites, independent price trackers and customer reviews — remains the safest route to a genuinely good deal.