Byte-Sized AI is a bi-weekly column that covers all things artificial intelligence—from startup funding, to newly inked partnerships, to just-launched, AI-powered capabilities from major retailers, software providers and supply chain players.
Meta wants other companies to use Business AI
Meta announced last week that it has aspirations to bring its AI-powered assistant, Business AI, to other companies.
The Big Tech giant said the change will allow companies to further personalize ads and product recommendations. It could also see consumers buying items directly via conversational AI, which major technology companies, like OpenAI and Amazon, are starting to double down on.
Related Stories
Business AI would enable third-party brands and retailers to provide stronger product information in a conversational format when users are served ads. According to CNBC, the feature will be free to incorporate into Meta ads, but if merchants choose to integrate the technology into their own sites, they’ll have to pay a fee that has yet to be determined.
CNBC reported that Clara Shih, the company’s head of business AI, told reporters during a press briefing that the move is aimed at making technology more broadly accessible to companies.
“We are going beyond ads and beyond Meta to help businesses drive impact across their customer experiences and customer operations,” Shih reportedly said.
Meta also announced last week that it would use the information users input into Meta AI, the chatbot it has built on its large-language model, Llama, to further personalize users’ experiences on its platforms. In a blog post, the company said that if a user had been discussing hiking with Meta AI, that user might then see a recommendation to join a hiking-focused Facebook group.
But the company could also use the data to target ads at consumers more effectively. In the hiking example, a user may be served ads for waterproof hiking boots on their Instagram feed after discussing hiking with Meta’s chatbot.
“Just like other personalized services, we tailor the ads and content you see based on your activity, ensuring that your experience evolves as your interests change,” the company said in a blog post.
The increased personalization based on Meta AI is set to take effect in mid-December.
Walmart leverages linguists and AI models to share stronger search results
Superstore giant Walmart has been working to help consumers navigate its product catalogue more efficiently. A major part of that strategy has involved leveraging AI models to ramp up the efficacy of its translation for digital search.
The company shared in mid-September that it continues to leverage AI, paired with the expertise of linguists in specific geographic areas, to better capture customer intent across languages.
The company first launched the tool, which it calls the Walmart Translation Platform, in 2022, starting with Spanish. Since, it has been adding to the experience for non-English speakers in a number of languages and specific geographies.
For instance, while many people in both Mexico and Chile speak Spanish, the specificities of the dialogue in both places differs. The WTC is designed to help bridge that gap and show users products that align with their initial search, rather than returning irrelevant results to the user.
“Sometimes, things that sound simple actually aren’t. Such is the case in translating the humble t-shirt. In Spanish, you might say camiseta, and expect it to be widely understood. But regional variations change how people describe the item,” the company wrote in a blog post. “It’s the soda is pop is Coke spectrum of middle America: In Mexico, people are likely to say playera, while in Chile you’re more likely to hear polera. Each a different word for the same thing. The WTP understands every one of these nuances and maps the variations back to the same entry, so customers find what they’re looking for — no matter where they’re searching from.”
What users see is largely powered by AI—but on the backend of the system, humans are helping the models learn.
Anna Lavinia Dambrosio, who oversees the team of linguists Walmart calls “localization experts,” said that synergy empowers customers across Walmart’s many markets.
“We really complement each other: the linguists train the model to be more human, while data scientists make it fast, scalable and technically efficient. We’re teaching AI to think like a human translator, where tone, grammar, context and intent can flip a sentence from technically correct to truly right for the customer,” she said in a statement. “It’s the partnership between human insight and machine learning that makes the platform unique.”
Alibaba and Nvidia want to work together to advance physical AI
Alibaba and Nvidia announced at the end of last month that they will collaborate on the proliferation of physical AI.
Physical AI helps autonomous machinery, like warehouse robots or driverless trucks, to better understand and work with the complexities of the physical world. It often combines information ingested by cameras or sensors with algorithms to help these types of machines with decisioning.
As autonomous robots rise in relevance and companies work to implement them across supply chain and logistics use cases, their ability to adapt on the go has become more important. Physical AI help robots, autonomous vehicles and other machines become more efficient, but it also works to ensure better safety outcomes for the humans in the loop.
Nvidia’s physical AI solutions will be embedded into Alibaba Cloud, which is Alibaba’s cloud computing platform, somewhat akin to Amazon Web Services (AWS). That means Nvidia’s proprietary embodied AI development kit, which allows companies to integrate AI systems into physical machinery so that they can effectively interact with the physical world.
The South China Morning Post reported that Alibaba Cloud unveiled the partnership at Apsara, its conference in Hangzhou, China.
AWS has also inked a partnership with Nvidia in recent months. The technology giant’s continued interest in making its models available across the industry could signal that further development for autonomous mobile robots (AMRs), including bipedal humanoid robots, is on the way.
