Listen to this story
|
On Thursday’s Q2 2023 earnings call, Amazon CEO Andy Jassy revealed that “Every single one” of Amazon’s businesses has “multiple generative AI initiatives going right now”. Founded as a retail store company, Amazon has been more focused on developing other AI-based products than its flagship e-commerce search in the recent past.
Over the past few months, the company has made several announcements and investments to accelerate through generative AI. During the AWS Summit in New York last week, the company launched Agents for Bedrock, for companies to use their own data to build AI apps that can automatically do tasks on their own foundational models, like image-to-text models or large language models, instead of just telling them about it. AI agents are the assistants which can actually do the job instead of just giving suggestions. For instance, an agent will book a flight instead of just giving a list of the cheapest available options.
Other AWS announcements include bringing generative AI to healthcare through AWS HealthScribe and a partnership with Nvidia allowing AWS to handle larger amounts of memory and data by using Nvidia H100 Tensor Core GPUs.
Subscribe to our Newsletter
Join our editors every weekday evening as they steer you through the most significant news of the day, introduce you to fresh perspectives, and provide unexpected moments of joy
Your newsletter subscriptions are subject to AIM Privacy Policy and Terms and Conditions.
Last month, the Bezos-owned company also made an announcement to invest $100 million in a AWS Generative AI Innovation Center, to help customers build and deploy generative AI solutions.
“Alexa, Reboot Yourself”
While the company has been internally transforming its AI force, they have also been hiring for Alexa AI from candidates with background in dialogue systems or information retrieval. Earlier this year in May, based on a leaked document titled “Alexa LLM Entertainment Use Cases,” the big tech plans to reboot Alexa the voice assistant with ChatGPT-like features. An internal memo described a goal of making the voice assistant smarter, stating users should feel “like Alexa is thinking vs. fetching from a database.”
These job posts clearly state that Amazon intends to overhaul its venerable voice assistant search feature. The current SVP of Amazon Devices & Services David Limp announced last month on LinkedIn that it will be hosting an event to presumably launch new devices on September 20th at the new HQ2 in Arlington, VA.
The partnership between HuggingFace and AWS gives further confidence that Amazon has something up its sleeve to boost users’ conversational experience with Alexa.
Looking out for Search
While the company’s voice assistant took Silicon Valley by a storm, the company’s bread and butter since day one has been e-commerce. Even though the retail giant’s search experience has been long criticised for ads bombarding the result page, the company is finally concerned about the feature.
Earlier this year, the company had also posted job listings for a machine learning-focused engineer describing how it is “reimagining Amazon Search” with a new “interactive conversational experience” to answer product questions, compare products, personalise suggestions, and more.
In June, the retail giant began testing a feature in its shopping app that uses AI to summarise reviews left by customers on some products. It provides a brief summary of what users liked and disliked about the product, along with a disclaimer that the overview is “AI-generated from the text of customer reviews.”
The recent frenzy around generative AI and chatbots like OpenAI’s ChatGPT, Google’s Bard and Microsoft’s Bing has pushed Amazon to sharpen its focus on the technology. The company which has been able to sell anything and everything since day one is now more involved in the AI field than ever. Across all verticals from Alexa to Code Whisperer, it is trying to step up its game. In the big picture, the IT giant is hustling with its generative AI agenda to keep up with its rivals’ efforts to hold the users’ attention.