Close Menu
Technology Mag

    Subscribe to Updates

    Get the latest creative news from FooBar about art, design and business.

    What's Hot

    Razer’s new Pokémon collab is not very effective

    July 17, 2025

    What Big Tech got out of Trump’s Big Beautiful Bill

    July 17, 2025

    I Thought the Dell 14 Plus Was Mid Until the Price Dropped This Much

    July 17, 2025
    Facebook X (Twitter) Instagram
    Subscribe
    Technology Mag
    Facebook X (Twitter) Instagram YouTube
    • Home
    • News
    • Business
    • Games
    • Gear
    • Reviews
    • Science
    • Security
    • Trending
    • Press Release
    Technology Mag
    Home » Former Top Google Researchers Have Made a New Kind of AI Agent
    Business

    Former Top Google Researchers Have Made a New Kind of AI Agent

    News RoomBy News RoomJuly 16, 20253 Mins Read
    Facebook Twitter Pinterest LinkedIn Reddit WhatsApp Email

    A new kind of artificial intelligence agent, trained to understand how software is built by gorging on a company’s data and learning how this leads to an end product, could be both a more capable software assistant and a small step toward much smarter AI.

    The new agent, called Asimov, was developed by Reflection, a small but ambitious startup cofounded by top AI researchers from Google. Asimov reads code as well as emails, Slack messages, project updates, and other documentation with the goal of learning how all this leads together to produce a finished piece of software.

    Reflection’s ultimate goal is building superintelligent AI—something that other leading AI labs say they are working toward. Meta recently created a new Superintelligence Lab, promising huge sums to researchers interested in joining its new effort.

    I visited Reflection’s headquarters in the Williamsburg neighborhood in Brooklyn, New York, just across the road from a swanky-looking pickleball club, to see how Reflection plans to reach superintelligence ahead of the competition.

    The company’s CEO, Misha Laskin, says the ideal way to build supersmart AI agents is to have them truly master coding, since this is the simplest, most natural way for them to interact with the world. While other companies are building agents that use human user interfaces and browse the web, Laskin, who previously worked on Gemini and agents at Google DeepMind, says this hardly comes naturally to a large language model. Laskin adds that teaching AI to make sense of software development will also produce much more useful coding assistants.

    Laskin says Asimov is designed to spend more time reading code rather than writing it. “Everyone is really focusing on code generation,” he told me. “But how to make agents useful in a team setting is really not solved. We are in kind of this semiautonomous phase where agents are just starting to work.”

    Asimov actually consists of several smaller agents inside a trench coat. The agents all work together to understand code and answer users’ queries about it. The smaller agents retrieve information, and one larger reasoning agent synthesizes this information into a coherent answer to a query.

    Reflection claims that Asimov already is perceived to outperform some leading AI tools by some measures. In a survey conducted by Reflection, the company found that developers working on large open source projects who asked questions preferred answers from Asimov 82 percent of the time compared to 63 percent for Anthropic’s Claude Code running its model Sonnet 4.

    Daniel Jackson, a computer scientist at Massachusetts Institute of Technology, says Reflection’s approach seems promising given the broader scope of its information gathering. Jackson adds, however, that the benefits of the approach remain to be seen, and the company’s survey is not enough to convince him of broad benefits. He notes that the approach could also increase computation costs and potentially create new security issues. “It would be reading all these private messages,” he says.

    Asimov deploys inside of customers’ virtual private clouds, so that all the data is retained by the customer.

    Share. Facebook Twitter Pinterest LinkedIn WhatsApp Reddit Email
    Previous ArticleElon Musk teases AI anime boyfriend based on Edward Cullen
    Next Article Elon Musk Unveils Grok 4 Amid Controversy Over Chatbot’s Antisemitic Posts

    Related Posts

    Elon Musk Unveils Grok 4 Amid Controversy Over Chatbot’s Antisemitic Posts

    July 16, 2025

    I Tried Grok’s Built-In Anime Companion and It Called Me a Twat

    July 16, 2025

    Thinking Machines Lab Raises a Record $2 Billion, Announces Cofounders

    July 16, 2025

    Tech Billionaires Back Erebor in the Wake of Silicon Valley Bank Collapse

    July 15, 2025

    Join Our Livestream: Inside the AI Copyright Battles

    July 15, 2025

    Microsoft and OpenAI’s AGI Fight Is Bigger Than a Contract

    July 14, 2025
    Our Picks

    What Big Tech got out of Trump’s Big Beautiful Bill

    July 17, 2025

    I Thought the Dell 14 Plus Was Mid Until the Price Dropped This Much

    July 17, 2025

    The next batch of emoji includes Bigfoot

    July 17, 2025

    Lenovo Legion Go S review part two: you were the chosen one!

    July 17, 2025
    • Facebook
    • Twitter
    • Pinterest
    • Instagram
    • YouTube
    • Vimeo
    Don't Miss
    News

    Wacom’s new MovinkPad drawing tablet doesn’t need a PC

    By News RoomJuly 17, 2025

    Wacom has announced the MovinkPad 11, an all-in-one Android-powered tablet for digital illustrators who want…

    AMD’s new 96-core Threadripper CPU will set you back $11,699

    July 17, 2025

    Perplexity’s CEO on why the browser is AI’s killer app

    July 17, 2025

    Dyson’s $500 headphones come in even more colors

    July 17, 2025
    Facebook X (Twitter) Instagram Pinterest
    • Privacy Policy
    • Terms of use
    • Advertise
    • Contact
    © 2025 Technology Mag. All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.