IT News Roundup: OpenAI Copyright Dispute, Meta Compute Cloud, and AI Model Shifts - July 14, 2026

The technology landscape this week has been dominated by escalating legal battles in the AI sector, major shifts in cloud infrastructure strategy, and a wave of significant cybersecurity threats. From publishers demanding sanctions against OpenAI to Meta entering the cloud provider arena, the AI industry continues to reshape itself rapidly. Meanwhile, security researchers have uncovered dangerous new exploits and botnets that pose immediate risks to organizations and home lab administrators alike.

New York Times-Led Group Demands Sanctions Against OpenAI in Copyright Trial

The New York Times, the Daily News, and other media outlets have asked a federal judge in Manhattan to impose sanctions on OpenAI, escalating one of the most consequential copyright disputes in AI history. The publishers allege that OpenAI has been misleading the court about its ability to search training datasets and ChatGPT output logs for evidence of their copyrighted articles.

In their motion, the publishers argued that OpenAI chose obstruction over transparency, failing to produce evidence at the start of the case that could have focused proceedings on the merits of its fair use defense. The case could set a major precedent for how AI companies handle copyrighted material in training data, with implications extending far beyond the immediate parties.

Source: Reuters, TechCrunch

Meta Launches Cloud Business to Monetize Excess AI Compute Infrastructure

Meta Platforms is developing a cloud infrastructure business under the internal name "Meta Compute" that will sell access to AI computing power and models, directly competing with Amazon Web Services, Google Cloud, and Microsoft Azure. The announcement sent Meta shares up nearly 9% as investors saw an opportunity to recoup some of the billions invested in AI data center infrastructure.

The service will offer businesses access to both raw compute capacity and Meta's AI models, including its recently launched closed-weight model Muse Spark. The move mirrors strategies employed by other tech giants and even SpaceX, which has explored monetizing excess satellite bandwidth. For homelabbers and IT professionals, this signals a new competitive dynamic in the cloud market that could eventually drive down AI inference costs.

Source: Bloomberg, TechCrunch

Anthropic Restores Claude Fable 5 Globally After US Export Controls Lift

Anthropic has redeployed Claude Fable 5 globally after the US government lifted export controls that had suspended access since June 12. The controls had covered both Fable 5 and its cybersecurity-focused counterpart, Claude Mythos 5. Fable 5 returned to users worldwide on July 1, with Mythos 5 access restored to approved US organizations.

Anthropic trained a new safety classifier to address the bypass technique that triggered the original export control order, reporting that the classifier blocks the technique in over 99% of cases. The company also announced it is extending Claude Fable 5 access on all paid plans through July 19, while keeping Claude Code's weekly rate limits 50% higher during the same period.

Source: VentureBeat, Forbes

OpenAI Consolidates Power Under Greg Brockman Ahead of Prospective IPO

OpenAI co-founder and president Greg Brockman has absorbed control of ChatGPT's product business, go-to-market operations, enterprise initiatives, and compute projects following the departure of Fidji Simo. The organizational consolidation centralizes authority under Brockman as the company prepares for a potential initial public offering.

The reshuffle merges ChatGPT, Codex, and the API into a single team under Brockman's leadership, sending a clear signal to investors about OpenAI's product strategy. The move comes amid ongoing legal challenges including Elon Musk's lawsuit over OpenAI's for-profit conversion and the escalating copyright dispute with major publishers.

Source: CNBC

Attackers Exploit SimpleHelp Zero-Day (CVE-2026-48558) to Deploy TaskWeaver and Djinn Stealer

Security researchers have documented active exploitation of a critical vulnerability in SimpleHelp, a remote support application, to deliver the TaskWeaver backdoor and Djinn Stealer malware. The zero-day, rated CVSS 10.0, allows unauthenticated remote code execution and has been actively weaponized against managed service providers and their clients.

The attack chain targets credentials across cloud platforms, code repositories, AI tools, and cryptocurrency wallets. Organizations using SimpleHelp or similar remote support tools should verify patch status immediately and monitor for signs of compromise including unusual outbound connections and credential access patterns.

Source: The Hacker News, DIESEC

RustDuck Botnet Rebuilds in Rust, Targeting IoT Devices for DDoS Attacks

Researchers at QiAnXin's XLab have been tracking RustDuck, an evolving DDoS botnet that has rewritten its codebase in Rust to improve performance and evasion capabilities. The malware hijacks routers, IP cameras, Android set-top boxes, and exposed servers since its emergence in February 2026.

RustDuck employs advanced encryption, anti-analysis evasion techniques, and exploits known IoT vulnerabilities to recruit devices into its network. The shift to Rust follows a pattern seen with RustoBot in 2025, indicating that threat actors increasingly favor modern, memory-safe languages for malware development. Home lab administrators should ensure IoT devices have firmware updates applied and unnecessary services are disabled.

Source: Security Affairs, iSecurify

Black Duck 2026 OSSRA Report: Open Source Vulnerabilities Double to 581 Per Codebase

Black Duck's 2026 Open Source Security Risk Analysis (OSSRA) report reveals that the average number of open source vulnerabilities per codebase has doubled to 581, representing a 107% year-over-year increase. The report found that 87% of codebases are at risk, with 65% having been hit by attacks.

The surge in vulnerabilities correlates with the rapid adoption of AI-assisted development tools, which introduce additional dependencies and increase the attack surface. The report highlights licensing conflicts and supply chain security as growing concerns alongside raw vulnerability counts. Organizations relying on open source components should implement automated dependency scanning and establish clear remediation SLAs.

Source: Black Duck