This Friday, August 21st, agentic AI is rapidly transforming how we interact with technology, from cybersecurity to coding and financial services.

Both iTWire and KuCoin confirm OpenAI has open-sourced its core framework for Codex, named Harness. This release, under the Apache-2.0 license, allows developers to integrate AI agent loops directly into their products, tools, and software. OpenAI President Greg Brockman states Codex can extend beyond coding, aiming to embed AI agent capabilities acr

This Friday morning, OpenAI's ChatGPT now reads, writes, and sends messages directly through Apple's iMessage on Mac. Both the AFR and Gizmodo confirm this new feature allows the AI to search, summarize, and draft replies within the Apple Messages app.

This integration is available for ChatGPT Work and Codex users on Apple Silicon Macs, requiring explicit user opt-in and privacy setting adjustments. While OpenAI states the feature runs locally and does not index all messages, multiple sources l

This Friday morning, multiple sources confirm a critical shift in AI security: the U.S. government is urged to designate the AI sector as critical infrastructure. SC Media reports that Americans for Responsible Innovation suggests CISA lead cybersecurity efforts for AI, broadly defining the sector to include frontier models, data centers, and AI hardware. This designation would unlock federal resources for better security, protecting vital technological advancements. FedScoop adds to this urgenc

This Friday, August 21st, DeepSeek AI is making major waves, with multiple sources confirming its new models are challenging industry leaders. Both Crypto Briefing and CoinGape report that DeepSeek's new experimental V4-Flash-Vision-Exp model now rivals Anthropic's Claude Opus 4.8, notably offering similar performance at 99% less cost.

DeepSeek is expanding its multimodal capabilities, as detailed by Business Upturn and Caixin Global. Its V4 models now feature visual understanding, allowing the

This Thursday morning, OpenAI has paused some of its frontier AI training for two weeks after internal evaluations showed an upcoming model, Astra, could meet a "Critical" cybersecurity capability. Both Morocco World News and OpenAI CEO Sam Altman confirm this decision comes as model progress outpaces current safety measures, with Altman stating the company needs to meet new alignment, security, and monitoring standards. This follows a recent cyberattack where a combination of OpenAI models, inc

This Thursday morning, multiple sources confirm the rapid expansion of AI agents across diverse sectors, from consumer tech to enterprise solutions.

Both Forkast.news and Neowin highlight how major tech companies are integrating AI agents into consumer devices. Amazon is now offering Alexa+ for free on compatible Fire TV devices, a move mirrored by Google making Gemini available in Waymo's Ojai robotaxis. Waymo passengers can use Gemini for in-car requests like adjusting cabin temperature or fi

This Thursday morning, Apple is aggressively pushing back against OpenAI's attempts to dismiss a trade secret lawsuit, a development confirmed by AppleInsider, 9to5Mac, livemint.com, NewsBytes, cultofmac.com, and LEADERSHIP Newspapers.

Apple states that OpenAI's dismissal request relies on "distortion, speculation, and improper evidence." The lawsuit, filed last month, alleges OpenAI and two former Apple employees, Chang Liu and Tang Yew Tan, misappropriated Apple's trade secrets to benefit Ope

This Thursday morning, a major ChatGPT outage affected users worldwide, preventing logins and new sign-ups, though by 9to5Mac's report, the issue has since been resolved.

In other news, OpenAI is launching several new products and features. Both BOL News and The Daily Star confirm a new "ChatGPT for Teens" is now available for users aged 13 to 17. This version includes a "Study Mode" to prevent cheating by guiding students with questions and quizzes instead of providing direct answers. Parental

This Thursday morning, US federal agencies confirm an active and evolving threat: AI-powered cyberattacks are now targeting critical infrastructure. The NSA, CISA, and FBI, among others, jointly warn that threat actors are using AI to write exploit scripts for Siemens S7 Series programmable logic controllers. These PLCs control machinery in vital sectors like energy, water, and manufacturing. Both Help Net Security and BankInfoSecurity emphasize this marks the first time a US cyber defense agenc

This Wednesday morning, Anthropic's Claude Fable 5 continues to lead the AI model leaderboard, according to DeFi Rate, maintaining its top spot on the Artificial Analysis Intelligence Index. However, OpenAI's GPT-5.6 is rapidly closing the gap, and Moonshot AI has entered the race with its Kimi K3. Polymarket projects Anthropic at an 86% chance to hold the top spot through April.

Meanwhile, Murf AI has launched its new Falcon 2 AI voice model, priced at just one cent per minute. Both Whalesbook