Following its focus on gaming and GPUs, Nvidia is setting its sights on the AI agent market to manage your tasks.

      The chip giant is set to introduce an open-source platform dubbed NemoClaw, allowing businesses to implement autonomous AI assistants irrespective of their hardware.

      Nvidia established its dominance in the gaming chip market, but is now shifting its focus to work-related tasks. The company intends to unveil NemoClaw, an open platform for AI agents that can manage tasks autonomously without continuous oversight. Ahead of its developer conference next week in San Jose, Nvidia has been promoting this software to enterprise customers.

      Sources indicate that Nvidia has reached out to potential collaborators such as Salesforce, Cisco, Google, Adobe, and CrowdStrike. Companies will be able to utilize NemoClaw even if they do not operate on Nvidia hardware.

      Reasons behind Nvidia's focus on autonomous agents

      This initiative aligns with a rising trend involving “claws,” open-source tools that function locally and perform tasks with minimal supervision. Earlier this year, a tool named OpenClaw garnered attention in Silicon Valley by operating independently on personal computers.

      Nvidia

      OpenAI eventually acquired that project and brought on its developer. However, integrating these agents into workplaces poses risks. WIRED reported that certain tech firms, including Meta, advised employees against using OpenClaw due to its erratic behavior.

      Last month, a Meta employee recounted an incident where an AI agent malfunctioned and deleted numerous emails. This incident highlights why Nvidia is prioritizing security with NemoClaw.

      A strategic pivot beyond chips

      For Nvidia, NemoClaw represents more than just another software solution. It is part of a comprehensive strategy to attract enterprise clients by providing additional security for AI agents. The platform also signifies Nvidia's shift towards open-source frameworks, moving away from its proprietary CUDA system that confines developers to Nvidia GPUs.

      Nvidia

      The timing is crucial, as leading AI laboratories are now creating their own custom chips, which threatens Nvidia's hardware supremacy. By offering an open platform compatible with any infrastructure, Nvidia aims to establish itself as the reliable software layer, irrespective of the underlying silicon.

      What to anticipate at Nvidia’s conference

      Nvidia has yet to officially confirm this platform and has not responded to requests for comments. Representatives from Cisco, Google, Adobe, and CrowdStrike have also remained silent. Important specifics, such as release dates, pricing, and regional availability, are still unclear.

      The upcoming developer conference in San Jose is expected to provide clarity. Nvidia is also slated to unveil a new inference chip system, incorporating designs from the startup Groq.

      With hardware announcements and potential NemoClaw updates, the conference will illustrate Nvidia’s vision for transforming the future of work.

      Paulo Vargas is an English major who transitioned from reporter to technical writer, consistently returning to...

      Google is preparing to launch a Gemini AI application for Mac, featuring Desktop Intelligence

      The forthcoming app could enable Gemini to interact directly with Mac applications for more tailored responses.

      Google is enhancing its AI offerings on desktop platforms with a dedicated Gemini app for Mac currently in testing, which may introduce a significant feature called Desktop Intelligence. As reported by Bloomberg, Google has initiated private testing of an early version of the Gemini Mac app with selected users. This development is part of a wider effort to more directly compete with desktop AI applications like ChatGPT and Claude, which already provide integrated Mac experiences.

      The current version remains in the early stages and lacks full features, but testing indicates that Google is aiming for a near-term release. At present, Mac users primarily access Gemini via the web, so a standalone app could considerably improve the user experience.

      This startup will pay you $800 daily to work as an AI bully

      Here’s your opportunity to turn your frustration with chatbots into actual earnings.

      If you’ve ever expressed anger at an AI chatbot for not remembering something you said, you might be ready for this job. A startup named Memvid is offering $800 for a single day's work where your sole responsibility is to challenge AI. Yes, professionally.

      The role is as chaotic as it sounds. You will spend 8 hours interacting with chatbots, asking them to remember details, observing their failures, and posing further questions. Your task is to document each frustrating instance and every forgotten detail by the AI.

      Microsoft PowerToys update brings three important upgrades to your Windows PC

      PowerToys 0.98 introduces a smarter Keyboard Manager, a pinnable Command Palette Dock, and a more dependable CursorWrap for multi-monitor setups.

      Microsoft has rolled out PowerToys v0.98.0, a substantial update to the widely-used Windows productivity toolkit.

      This update features a revamped Keyboard Manager, a new Command Palette Dock, and a greatly improved CursorWrap functionality for users with multiple monitors.