Amazon's AI wants to own online shopping data
The goal? Make every listing on Amazon accurate, detailed and easy to understand, whether the product is sold by Amazon or a third-party seller. If the project works as planned, it could save sellers hours of work and help shoppers find what they need faster.
Sign up for my FREE CyberGuy ReportGet my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox. Plus, you'll get instant access to my Ultimate Scam Survival Guide — free when you join my CYBERGUY.COM/NEWSLETTER
Starfish is a multi-year initiative built around generative AI. According to an internal Amazon document obtained by Business Insider, the system gathers product data from across the web, including external websites and images. It then uses large language models (LLMs) to create "complete, correct and consistent" product listings. This isn't a small update. Amazon expects Starfish to boost sales by $7.5 billion in 2025 alone by improving conversion rates and expanding product variety.
Starfish builds on earlier AI tools that Amazon began testing in 2023. These tools could:
Now, with Starfish, Amazon wants to scale that effort across millions of listings. The AI will also collect data from 200,000 external brand websites by crawling, scraping and mapping their content to Amazon's catalog. It's not yet clear whether Amazon's own web crawler, Amazonbot, is powering Starfish. But the company confirmed to Business Insider that Starfish is already supporting its new "Buy for Me" feature. This feature recommends products from external websites and lets shoppers buy them directly within Amazon's app.
Manually creating product listings is slow and often inconsistent. That's a problem when Amazon wants to offer a massive selection with reliable information. If shoppers can't find what they're looking for, or if the listings are vague, they may head elsewhere. Starfish addresses this by automating the tedious parts of listing creation. That helps sellers spend less time writing and more time selling. For Amazon, better listings mean higher conversion rates and happier customers. Plus, this move positions Amazon to compete more directly with Google Shopping, which also aims to be a central hub for product information.
Amazon is testing Starfish's effectiveness with A/B comparisons, measuring sales performance of AI-enriched listings versus standard ones. It's also rolling out bulk listing tools and preparing to expand the system globally. This isn't just about improving Amazon's website. It's about changing the way product information is gathered, created and shared at scale.
If you're a shopper on Amazon, this could mean faster access to clearer, more accurate product listings, especially for obscure or hard-to-find items. As Amazon's AI fills in missing details and improves titles and descriptions, the results should help you make better decisions with less research.
For sellers, this streamlines the work of creating listings. If you've struggled to write compelling descriptions or keep up with Amazon's catalog standards, the Starfish project may do much of the heavy lifting. That could save time, reduce errors and improve sales performance.
However, there are some trade-offs. As Amazon scrapes more data from across the web to power its listings, brands and smaller websites may worry about how their product information is being used. And if AI-generated content becomes widespread, quality and trust in listings may vary depending on how well the system works.
In short, expect a more automated Amazon shopping experience, with both conveniences and questions about how your data and the broader web are being used to power it.
Amazon's Starfish project signals a major shift in how e-commerce works. By combining web scraping, AI models and deep integration into its Marketplace, Amazon hopes to automate one of the most time-consuming parts of online selling. For buyers and sellers, this could mean more convenience and better results. But it also raises important questions about transparency, data ownership and the future role of AI in shaping what we see online.
Would you trust AI to tell you everything you need to know before you click "Buy Now"? Let us know by writing us at Cyberguy.com/Contact
Sign up for my FREE CyberGuy ReportGet my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox. Plus, you'll get instant access to my Ultimate Scam Survival Guide — free when you join my CYBERGUY.COM/NEWSLETTER
Copyright 2025 CyberGuy.com. All rights reserved.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Forbes
a few seconds ago
- Forbes
5 Hacks To Quickly Boost Your Google Traffic With AI In 2025
To avoid being outnumbered or outranked by AI, you've got to use the same playbook that AI follows Your website's Google traffic isn't down because the algorithm hates you. It's down because SEO as you once knew it, is dead. Welcome to SGE, the new face of SEO. This is where you get more website visitors, higher rankings, and more clients. In this article, you'll learn a highly likely reason why your web traffic is down this year, what you can do about it, and how, as a freelancer or content creator you can cooperate with the culprit (AI) to drive visibility and traction again. Why Is My Website Traffic Dropping? If you're a freelance writer, blogger, or content creator, and you notice the site traffic declining, or if you're a freelance marketing professional and you're observing your clients' traffic plummeting, these are the most likely reasons. What Is SGE And How Does It Affect SEO? SGE means Search Generative Experience, as opposed to SEO, which means Search Engine Optimization. It's also known as GEO, Generative Engine Optimization, where you're focused on appearing in AI search results. SGE is when you type something into Google Search, and instead of you scrolling through websites to find the perfect answer, the answer is positioned at the top of the results page in an AI Overview box. Google's AI Overview briefly summarizes main points as it presents your answer, and importantly, provides links to where its information is sourced. Why Traditional SEO Isn't Working Anymore Traditionally, digital marketing professionals use SEO principles and techniques to rank highly on Google and appear at the top of search results, which brings in more web traffic, and consequently, more sales. But if you're a content creator, freelancer, entrepreneur, or freelance copywriter, you'll likely find yourself running into a wall because no matter how hard you try, Google is de-ranking you. How Can I Increase My Google Traffic Using AI? The new Google algorithm updates and AI Overviews feature, as well as the use of AI-powered search engines and ChatGPT's built-in web search feature has many professionals furious. I noticed one LinkedIn member took to the platform to express their rage at their traffic dramatically shrinking for no apparent reason. Reddit users are also sharing their experiences, noting a fall in web traffic for their websites and blogs, due to AI overviews. However, vice president of marketing and start-up advisor Andrea Palten recommends: 'If you want to show up in AI answers, start publishing content that's quotable, useful, and easy to scan.' Palten suggests these hacks for ensuring AI tools like Perplexity and Google's AI Overview feature your site: These make it easy for AI to parse your work. To stand out with your thought leadership online, your post/article/blog can't replicate everyone else's or be generic. Never entirely create content using AI tools like ChatGPT, because it becomes too general, there are no fresh insights, and it sounds mediocre. Instead, try this: This boosts your authority and visibility online as an industry expert and thought leader. One of the best formats is FAQ-style writing. This makes your blog content more likely to be quoted directly in AI-generated responses. AI tools often pull from structured, numbered content. Try titles like: Make sure your name or brand appears somewhere in the post (and ideally in the title image or author box). These posts are easy for AI to summarize, and cite. Don't waste time rambling with a lengthy introduction. To establish thought leadership and get more clients, focus on creating content that is easily ... More scannable Using these AI hacks, you're able to work with AI, and not view it as your enemy. I've personally tried these suggestions and have seen tremendous results in the viewership of my content online, even at a time when overall traffic is dying across multiple platforms. In a nutshell: "Write clearly. Publish often. Make it easy to cite. "That's how you get found," Palten says.
Forbes
a few seconds ago
- Forbes
Meeting The AI Era Power Demand Through Collaboration
Javier Cavada, President and CEO of EMEA at Mitsubishi Power. Every time we ask ChatGPT a question, stream our favorite series or look up directions on our phone, we're tapping into a surge of electricity that's quietly reshaping our world. Behind the screens and servers, we are witnessing the expansion of data centers at an incredible rate—and with them, our skyrocketing appetite for power. Recent reports, including one from the International Energy Agency (IEA), project that energy demand from AI-driven data centers is set to quadruple globally by 2030. This is echoed in the U.S. and regions across the world. In the UK, data center electricity consumption is forecast to rise from 2.5% of total usage today to nearly 6% by 2030. Across Europe, the rise is expected to more than triple. These are signs of a shift already underway. The digital economy is growing exponentially, and it needs reliable power that can keep up. What's compounding the situation is that this surge in demand is happening just as the global community is doubling down on climate goals. So, the real question becomes: How will we keep pace with the power needs of the future without compromising our progress toward Net Zero? Where We've Been And Where We're Heading As we transition to energy systems that are majority renewable-powered, the grid requires more support and storage to balance supply and demand from wind and solar, which are intermittent. Gas provides what we call dispatchable power—power that can be dialed up and down almost instantaneously to support intermittent forms of power. But, as my work with power operators has shown me, electricity is growing as a share of the overall power mix as the electrification of power, heat and transport increases. Electricity, once just another utility, is now central to everything. It's becoming the foundation of all infrastructure—not just digital but also transportation, manufacturing, and the heating and cooling of homes and cities. And as that happens, electricity is becoming a much bigger share of global emissions. That's why it's important that how we generate power evolves as fast as how we consume it. The recent power outage in Spain and Portugal demonstrated that we live in an increasingly interconnected world. Simply put, we depend on many individuals to meet our power needs—and increasingly, people want those needs met with clean power. While technology developers have a central role, I believe it is only through partnerships that we can scale these technologies to enable widespread decarbonization. Delivering Climate-Responsible Power We don't necessarily have to choose between digital innovation and climate responsibility; if we get it right, we can achieve both. Of course, there are challenges to this. The growth of data centers, especially those driving AI, brings unique demands. These facilities can't afford a second of downtime. Cooling systems must run continuously, computing loads are relentless, and every backup needs a backup. That kind of unbreakable reliability comes only from a well-balanced energy system. One impactful change we can make today is modernizing fossil-fuel-based infrastructure. Around 16% of Europe's electricity still comes from coal-fired power plants—and in some countries, it's higher. Replacing coal with modern, high-efficiency gas turbines could cut emissions by more than 60%. This demonstrates the importance of moving toward power sources that are low-to-zero-carbon. However, if we want to scale the options available to us, I believe we need to set in motion a wider value chain that unlocks investment and de-risks growth. A major part of development is creating a supportive policy environment. As leaders, we all have a role to play to ensure innovation and technology are enabled to succeed—starting with our business operations. We don't need to wait for new inventions; we just need to implement what's already proven. In my experience, that typically happens through collaboration with partners like investors, policymakers, utilities, developers, off-takers and end-users. Business leaders fall into one or more of these groups—and they have the influence to act. Part of acting is taking responsibility within our own companies for questions like, "How is our power demand being met? By whom? And how clean is it?" The scale of this transition can feel overwhelming, but it's a historic opportunity. The rise of data centers and the energy they require is the perfect reason for us to double down on building the clean energy systems we need anyway. Placing Power Demand In Concert With Climate Needs Across Europe, power security is a major part of the conversation, and I've observed growing recognition that clean solutions must be part of the equation. But we need more clarity, consistency and urgency. Every data center built without a clean, resilient power plan is a missed opportunity. AI can help—not just in generating demand but in managing the grid efficiently, forecasting usage and optimizing dispatch. The more digital our infrastructure becomes, the more "intelligent" our energy systems must be. The demands of AI and the climate do not have to be in conflict; innovation in digital technologies can be matched by innovation in energy. More than just engineering better turbines or blending fuels, this is about powering a new era where reliability and sustainability go hand-in-hand. The electricity that powers our lives today should also protect our future. That's the challenge. But with the right choices, it can also be the solution. Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?
Yahoo
28 minutes ago
- Yahoo
Verizon boosts annual profit forecast on demand for premium plans
(Reuters) -Verizon raised the lower end of its annual profit forecast on Monday, as strong demand for its higher-tier plans powered better-than-expected earnings in the second quarter. The U.S. telecom major posted a 2.2% rise in wireless service revenue as more users opted for its add-ons such as access to streaming service like Netflix. The carrier has launched price-lock promotions and broadband-wireless bundles to retain users as competition intensifies from AT&T and T-Mobile, as well as aggressive offers from broadband providers Comcast and Charter. However, Verizon posted a surprise drop of 9,000 monthly bill-paying wireless subscribers in the April-June period, reeling from user churn after price hikes in January. Analysts polled by FactSet were expecting an increase of 13,000 subscribers. To drive growth in the mature U.S. telecom market, Verizon and its wireless rivals have been bulking up on fiber-optic assets that can tap growing data use by customers. Verizon in May won approval from the U.S. telecom regulator for its $20 billion acquisition of fiber-optic internet provider Frontier, after it agreed to end its diversity programs. The sharper focus on internet services helped it posted 293,000 broadband net additions in the second quarter. Overall, Verizon reported revenue of $34.5 billion, beating estimates of $33.74 billion, according to data compiled by LSEG. Its quarterly adjusted earnings per share of $1.22 also beat estimates. The company now expects 2025 adjusted profit to grow between 1% and 3%, compared with 0% to 3% previously. It also raised its annual free cash flow forecast to between $19.5 billion and $20.5 billion, from $17.5 billion to $18.5 billion.
