Meta to begin manufacturing 'Iris' AI chip in September as AI push gathers pace: Report

firstpost.com Jul 10, 2026

Meta to begin manufacturing ‘Iris’ AI chip in September as AI push gathers pace: Report

Curated by Shubham · TechCapsules · Jul 10, 2026

AI-summarised brief · reviewed before publication

Meta plans to commence manufacturing its custom “Iris” AI chip in September, marking a significant acceleration in its strategy to expand AI computing capacity to 14 gigawatts by 2027. According to an internal memo reviewed by Reuters, the chip, developed with Broadcom and produced by TSMC, has passed six weeks of testing without major issues. This progress revitalizes Meta’s in-house silicon initiative, which has faced challenges since launching over five years ago. The Iris chip is part of a four-generation roadmap designed to reduce reliance on external suppliers like Nvidia and AMD while lowering computing costs. Meta aims to release new AI chips every six months through 2027, a pace faster than industry norms. The company intends to deploy 7 gigawatts of infrastructure in 2026, doubling it the following year, supported by an estimated $145 billion annual spend on AI infrastructure.

💡 Why It Matters

· By shifting to a six-month release cycle, Meta challenges the traditional annual hardware cadence, forcing competitors to accelerate their own innovation timelines.
· This aggressive internal development strategy directly threatens the market dominance of Nvidia and AMD in the high-performance computing sector.

AI Capsules Meta Semiconductors

Read full article on firstpost.com

Xbox CEO Asha Sharma Is Advising the Federal Reserve on the Economic…

Xbox CEO Asha Sharma has joined a Federal Reserve taskforce to evaluate the economic impact of AI and other technologies on productivity and jobs. Sharma, who recently restructured the Xbox business, including cutting 1,600 jobs, brings experience from Microsoft’s AI division. She joins figures like Marc Andreessen and Stanford economist Charles I. Jones. The taskforce aims to assess how technological shifts affect employment and economic policy.

💡 Sharma’s dual role at Microsoft and the Fed highlights the growing intersection between corporate tech strategy and macroeconomic policy. Her insights into AI’s role in both job displacement and innovation could shape how central banks respond to future economic disruptions.

techcrunch.com · Jul 10

Elon Musk praises Mythos/Fable, promises not to ‘cut off’ Anthropic

Elon Musk praised Anthropic, vowing not to cut off the AI lab from SpaceX’s servers despite past rivalry. Anthropic, now SpaceX’s largest customer, signed a $40 billion deal to use SpaceX’s compute power through 2029. Musk cited Tesla and SpaceX’s open practices as proof of his fair competition stance. The partnership benefits both companies, with Anthropic gaining infrastructure and SpaceX learning from Anthropic’s AI growth.

💡 The deal signals a shift in Musk’s approach to competition, aligning business interests with a once-rival AI lab. By hosting Anthropic, SpaceX gains deep technical insight into one of the industry’s leading models, potentially accelerating its own AI ambitions.

techcrunch.com · Jul 10

OpenAI is shutting down Atlas, but its AI browser ambitions are still…

OpenAI is shutting down Atlas, its AI-powered browser launched in October with ChatGPT at its core. Instead of abandoning AI-driven browsing, the company is integrating Atlas’s features into ChatGPT’s desktop app and a Google Chrome extension. This shift follows OpenAI’s decision to cut back on non-core projects, including its AI video tool Sora. The move reflects a broader industry trend as companies like Google, Microsoft, and Perplexity compete to redefine how users interact with the web through AI-enhanced browsers.

💡 By embedding AI browsing tools into existing platforms like Chrome and its desktop app, OpenAI is positioning ChatGPT as a central hub for productivity, rather than a standalone browser. This strategy allows it to reach users where they already spend time, increasing the likelihood of long-term adoption over competing AI-integrated browsers.

thequantuminsider.com · Jul 10

Google Study Shows Quantum Computer Can Learn From Its Own Errors While…

Google researchers have demonstrated a reinforcement‑learning AI system that continuously calibrates a quantum processor while it runs, using the error‑correction data generated during computation. Tested on the Willow superconducting chip, the AI adjusted more than 1,000 control parameters in real time, cutting logical error rates by roughly 20 % compared with traditional calibration and keeping performance 3.5 times more stable when artificial hardware drift was introduced. By eliminating the need for periodic, offline recalibration, the method enables longer, uninterrupted quantum operations and sets new benchmark results for both surface‑code and color‑code error correction on superconducting hardware. The approach is presented as a step toward autonomous, fault‑tolerant quantum computers that can maintain precision over days‑long calculations.

💡 Continuous, AI‑driven calibration removes a major scalability bottleneck, allowing future quantum machines to run complex algorithms without costly downtime.

completeaitraining.com · Jul 10

Liberty Mutual modernizes legacy systems to scale AI use cases

Liberty Mutual unveiled a conversational AI auto‑insurance quoting app in May, allowing customers to obtain quotes via OpenAI’s ChatGPT instead of web forms. The rollout follows a decade‑long migration of 85% of workloads to the cloud, consolidation of 13 data centers into a facility and a hybrid strategy across AWS, Google and Azure. CIO Palmer highlighted an abstraction layer built with IBM that overlays mainframe data, using the open Model Context Protocol to keep the system agnostic and preserve context for complex business logic. The insurer is also replatforming core systems to Guidewire over five years and has created an AI FinOps team to monitor token consumption, linking spend to business outcomes. Palmer warned that without infrastructure, initiatives will falter.

💡 Decoupling AI models from legacy mainframe data lets insurers scale generative AI while avoiding vendor lock‑in, offering a practical roadmap for industry‑wide digital transformation.

en.vietnamplus.vn · Jul 10

Ho Chi Minh City launches pilot digital investment promotion infrastructure

Ho Chi Minh City unveiled a preview of HCMCInvest, a digital investment promotion platform, on July 9. The system uses artificial intelligence and data‑driven tools to centralise information on urban planning, infrastructure, land resources and priority sectors, aiming to remove information barriers and create a level playing field for domestic and foreign investors. Municipal People’s Committee Chairman Nguyen Van Duoc said the platform will shorten market‑research time, improve transparency and boost the city’s competitiveness, while acknowledging a cautious rollout with ongoing adjustments. The National Digital Transformation Agency’s director, Tran Duy Ninh, praised the city as a pioneer in integrating AI into investment promotion, aligning with global digital trends. Arobid.com and the city’s Investment and Trade Promotion Centre signed an agreement to manage and optimise the platform, establishing a monitoring mechanism.

💡 By digitising investment data, Ho Chi Minh City can accelerate decision‑making for investors and sharpen its edge in the regional race for foreign capital.