Al large language models are old hat, Meta Chief AI Scientist Yan LeCun told a private breakfast organized by the Joint European Disruptive Initiative (JEDI) during VivaTech, Europe’s largest tech conference. Focus on what’s next, he urged participants. “Every single interaction will be mediated by AI assistants,” personalized bots that help you work, create, or communicate better, and interface with the digital world on your behalf, he said.
Meta has already launched an AI assistant on its platforms Facebook, Instagram, and WhatsApp across more than a dozen countries. Google and its AI division DeepMind, as well as Microsoft-backed OpenAI, along with Microsoft itself and other tech companies, are all racing to offer AI personal assistants to everyone from consumers to business users to teachers.
In the past week Google and Open AI simultaneously announced a series of upgraded AI tools that are “multimodal”, which means they can interpret voice, video, images, and code in a single interface, and carry out complex tasks like live translations or planning a family vacation.
In a video demonstration, Google’s prototype AI assistant Astra, powered by its Gemini model, responded to voice commands based on an analysis of what it sees through a phone camera or when using a pair of smart glasses. Among other things it successfully identified sequences of code and reminded the user where they had left their glasses.
ChatGPT maker OpenAI said it would release a new AI model called GPT-4o, capable of realistic voice conversation and able to interact across text and image, its latest move to stay ahead in a race to dominate the emerging technology. New audio capabilities enable users to speak to ChatGPT and obtain real-time responses with no delay, as well as interrupt ChatGPT while it is speaking, both hallmarks of realistic conversations that AI voice assistants have found challenging, the OpenAI researchers showed at a livestream event. “The new voice (and video) mode is the best computer interface I’ve ever used,”OpenAI CEO Sam Altman said in a blog posting. “ It feels like AI from the movies.” The voice sounded so much like Scarlett Johansson playing an AI character in the sci-fi movie “Her” that the actress objected and OpenAI dropped the voice from its collection.
Meanwhile, Microsoft this week announced an upgraded version of Copilot, its AI assistant. The announcements, ahead of Microsoft’s annual Build developer conference in Seattle, centered on embedding AI features into a product where Microsoft already has the eyes of millions of consumers: the Windows operating system for personal computers. The new features will include Windows Recall, enabling the AI assistant to “access virtually what you have seen or done on your PC in a way that feels like having photographic memory”. Microsoft promises to protect users’ privacy by giving them the option to filter out what they don’t want tracked.
Microsoft also announced a new AI assistant for teams that keeps meetings on track, takes collaborative notes, manages large projects, and assigns tasks. The company noted that people are ultimately in control — with the ability to ignore or override the choices or assignments made by the Team Copilot. They can also turn the tables and assign tasks to the Team Copilot instead.
Microsoft is also taking steps to ensure that AI assistants are being used outside of the office. It announced May 21 that it is partnering with tutoring organization Khan Academy to provide a generative AI assistant to all teachers in the U.S. for free Khanmigo for Teachers, which helps teachers prepare lessons for class, is free to all educators in the U.S. as of Tuesday, with the financial support of Microsoft. The program can help create lessons, analyze student performance, plan assignments, and provide teachers with opportunities to enhance their own learning. Khan Academy has roughly 170 million registered users in over 50 languages around the world.
Apple is also expected to be a major player in this race. Industry insiders anticipate that a significant upgrade to Apple’s voice assistant, Siri, is on the horizon, as the company rolls out new AI chips, designed in-house and capable of powering generative models on-device.
During the May 23 breakfast meeting Meta’s LeCun talked about the many new opportunities AI assistants will offer for the tech industry. Along with actually creating the bots there will be a need to develop low power kits to run AI systems on eyeglasses or pocket devices, new chips to power the devices and new types of batteries, he said.
But, as with all new AI developments, the potential impact on society must be taken into consideration, says a Google DeepMind white paper on the ethics of advanced AI assistants, published on May 19.
“Our analysis suggests that advanced AI assistants are likely to have a profound impact on our individual and collective lives,” says the paper. “To be beneficial and value-aligned, we argue that assistants must be appropriately responsive to the competing claims and needs of users, developers, and society. Features such as increased agency, the capacity to interact in natural language and high degrees of personalization could make AI assistants especially helpful to users. However, these features also make people vulnerable to inappropriate influence by the technology, so robust safeguards are needed. Moreover, when AI assistants are deployed at scale, knock-on effects that arise from interaction between them and questions about their overall impact on wider institutions and social processes rise to the fore.”
The paper says these dynamics will “likely require technical and policy interventions” in order to to achieve broad, inclusive and equitable outcomes. Finally, given that the current landscape of AI evaluation focuses primarily on the technical components of AI systems, the paper says it is important to invest in “holistic sociotechnical evaluations of AI assistants, including human–AI interaction, multi-agent and societal level research, to support responsible decision-making and deployment in this domain.”
To access more of The Innovator’s News In Context stories click here.
