Cybersecurity

Pennsylvania's We Love Philly organization offers pre-apprenticeships in digital marketing and cybersecurity to high school students and recent graduates. The state has expanded these programs, with nearly 3,200 participants in 2023, and Governor Josh Shapiro proposes doubling investment to $7 million. However, there is no guarantee of jobs post-training, and data on program effectiveness remains limited.

An international intelligence coalition, Five Eyes, warns that AI is enabling more sophisticated cyberattacks targeting small businesses and local governments. Cybersecurity experts highlight the vulnerability of these entities due to limited resources and recommend immediate defensive measures against AI-driven threats.

The article discusses how security leaders struggle not with finding vulnerabilities but prioritizing them, as AI's acceleration of vulnerability discovery exacerbates the problem of overwhelming data without clear risk context. It highlights that organizations focus on visibility metrics rather than contextual risk assessment, leading to inefficient resource allocation.

Google's Wiz, after its $32 billion acquisition, has seen a surge in job applications. A talent leader at Wiz emphasized that candidates should align their experience with the company's current growth stage and focus on customer challenges rather than just the company's status.

Z.ai, a Chinese startup, launched ZCode, an AI coding tool competing with Cursor and GitHub Copilot, offering lower pricing tiers and integration with other models. The tool is part of Z.ai's open-source GLM 5.2 ecosystem, which previously challenged Silicon Valley with its performance in cybersecurity and high context windows.

NR Labs, a Tyson, Virginia-based cybersecurity firm founded by Brian Gay and Ryan Tappis, has been named a Top Workplace twice in its first two years. The company emphasizes a people-driven culture, prioritizing employee happiness and removing bureaucratic barriers to focus on core work.

Anthropic will redeploy Fable 5 following the US lifting export controls. The model will use new classifiers to block a broader range of cybersecurity-related tasks.

Palo Alto and CrowdStrike reported their best quarter ever due to increased demand driven by AI threats. Both companies are focusing on identity security, a critical segment in cybersecurity, as AI agents outnumber humans.

Chinese artificial intelligence models have matched top U.S. systems in cybersecurity, according to a report. A new model from China's Zhipu AI is now comparable to Anthropic's systems, potentially influencing U.S. AI policy discussions.

China's Zhipu AI released its open-weight GLM-5.2 model, which researchers claim matches Mythos in cybersecurity and bug-finding capabilities. The US government has restricted access to advanced models like Mythos and Fable, as well as training hardware, due to concerns over China's narrowing technological gap.

The Pentagon is advancing AI, cybersecurity, and cyber strategy to strengthen its future force, as detailed in an eBook based on Breaking Defense’s TechNet Cyber 2026 coverage.

OpenAI has released GPT-5.6 to select users vetted by the US government, offering a limited preview of new models with advanced cybersecurity capabilities.

The U.S. military's cybersecurity efforts began over 50 years ago, with early recognition in 1972 that computers could be targeted. A 1997 exercise, Eligible Receiver 97, exposed critical vulnerabilities, prompting major strategic changes and shaping the creation of U.S. Cyber Command.

Infosys chairman Nandan M. Nilekani argues that AI coding tools will not replace services companies like Infosys because software development requires more than just coding, such as enterprise context, cybersecurity, and integration with legacy systems. He claims AI will create demand for human expertise in modernizing outdated software and integrating AI with traditional enterprise platforms.

The Five Eyes spy agency alliance warns that advanced AI models are improving rapidly, potentially bypassing current cybersecurity measures within months.

OpenAI has launched Patch the Planet, part of its Daybreak cybersecurity program, to help open-source projects combat bugs.

The Five Eyes intelligence agencies from Australia, Canada, New Zealand, the USA, and the UK issued a warning that AI-powered cyberattacks could escalate into major operational and financial crises, urging organizations to prioritize foundational cybersecurity practices and integrate AI tools for defense. They emphasized that breaches are inevitable and preparedness is critical to mitigate risks, while acknowledging AI's dual role in accelerating threats and enhancing resilience.

Artificial intelligence is already integrated into national security operations, including analysis, cyber defense, and mission planning. The article highlights risks such as bad data, weak access controls, and untested automation, emphasizing the need for AI assurance to mitigate operational risks in sensitive environments.

A ransomware attack at Evanston Township High School underscores growing cybersecurity threats, as experts warn AI is making cyberattacks more sophisticated and harder to detect. The incident highlights concerns about AI's role in escalating online safety risks.

Digital agents are increasingly making online purchases, prompting companies to adapt their websites to serve these agents effectively. Platforms like CHEQ help businesses distinguish between bots and agents by analyzing user behavior and device data, with agents projected to handle $5 trillion in retail transactions by 2030.

Students and working adults increasingly prioritize faster, cheaper, and job-focused credentials over traditional four-year degrees. Data shows a rise in non-degree certificates in fields like healthcare and IT, with 3.4 million credentials awarded in 2024-25. A Washington Post example highlights a human resources executive earning a bachelor’s and master’s in months for under $4,000.

The article discusses the increasing sophistication and scale of cyberattacks, driven by AI-enhanced malware and a fourfold rise in daily attacks reported by Palo Alto Networks. Experts warn of vulnerabilities in internet security, with AI tools enabling faster and more complex hacking methods.

The US has imposed limits on Anthropic's Fable AI, which was designed for advanced cybersecurity work. The shutdown of Fable 5 highlights a dilemma in AI security, where the same tools can benefit both defenders and attackers.

The article argues that veterans are underutilized in the defense-tech sector despite their valuable experience with technology and operational environments. It highlights the mismatch between veteran transition programs and the growing demand for skills in AI, cybersecurity, and autonomous systems, noting over 200,000 annual military separations and a rapidly evolving tech landscape.

Vietnam's new government is prioritizing economic growth through partnerships with U.S. companies in technology, energy, and digital infrastructure. The government aims to reduce regulations and leverage U.S. expertise in AI, cloud computing, and renewable energy, while addressing challenges like data localization policies to attract investment.

Anthropic shut down access to its Mythos 5 and Fable 5 models following a US Commerce Department directive imposing export controls. The move was prompted by concerns over a jailbreak method bypassing safeguards for cybersecurity, chemistry, and biology-related prompts, leading the administration to request a pause in model releases to strengthen national security measures.

CyberCorps: Scholarship for Service program is adapting to AI-driven cybersecurity challenges but faces budget risks under the Trump administration. Congress has intervened to fund the program, which trains cybersecurity professionals with AI expertise to defend against evolving threats.

The European Union and Brazil are signing a digital partnership to collaborate on technology, aiming for Europe to reduce reliance on US tech giants. The cooperation will focus on data, internet connectivity, cybersecurity, and protecting children online.

US lawmakers emphasized the need for America to lead the next phase of AI innovation to prevent China from gaining a technological advantage, citing cybersecurity and national security concerns. They warned against adversaries like China surpassing the US in artificial intelligence.

Anthropic released Claude Fable 5, its most powerful publicly available AI model, but it avoids answering basic biology questions by redirecting to the older Claude Opus 4.8 model. The restriction is by design, as Fable is part of the Mythos-class models deemed too dangerous for cybersecurity tasks, though it retains the capability to answer such questions.

Anthropic is releasing a public version of its Mythos AI model, Claude Fable 5, with restrictions on cybersecurity applications. The model, limited to 200 organizations including the US government under the Glasswing program, is designed to avoid tasks like identifying software vulnerabilities. It will be more expensive but use fewer tokens, reducing overall costs.

Anthropic's Claude Fable 5 model, part of the 'Mythos-class' series, includes broad safety safeguards that can block or revert to Opus 4.8 when queries about cybersecurity, biology, or chemistry are flagged. The model's restrictions were implemented to prevent potential misuse in sensitive areas, even if they mistakenly block benign requests.

Anthropic released Fable 5, a new version of its Claude Mythos AI model to the public, while restricting access to its most powerful Mythos model due to cybersecurity concerns. The model is the first from the company’s new Mythos class, previously restricted to partner institutions for months.

Anthropic is releasing a new AI model, Claude Fable 5, to the public with safety measures limiting its use in cybersecurity and biological research. The model includes guardrails to prevent potential harm in these high-risk areas.

Anthropic has released a public version of its Mythos AI model, now named Claude Fable 5, with safeguards preventing use in cybersecurity and other risky areas. The model is described as the company's most powerful for wider use, excelling in software engineering and analytics, though access is currently limited to around 200 users.

Anthropic released Claude Fable 5, a 'Mythos-class' AI model with safeguards restricting cybersecurity, biology, chemistry, and distillation queries to its Opus 4.8 model. The model exceeds previous Anthropic models in capabilities but is limited in access, with a 'trusted access program' planned for a version without full safeguards.

Anthropic and OpenAI are implementing selective access programs for their advanced AI models, prioritizing cybersecurity applications while restricting high-risk capabilities. Anthropic's Fable 5 model routes sensitive requests to a less capable model, while OpenAI offers tiered access to its GPT-5.5 variant for cybersecurity researchers.

Anthropic is releasing a public version of its Mythos AI model as Claude Fable 5. The model costs $10 per million input tokens and $50 per million output tokens, with hard limits on cybersecurity and biology queries.

Anthropic released a guardrailed version of its Mythos model called Fable 5, which blocks answers to cybersecurity and biology-related questions to ensure public safety. The company claims its safeguards withstood hacker testing, though concerns remain about potential jailbreaking attempts. Anthropic also upgraded Mythos 5 for select customers, highlighting its strong cybersecurity capabilities and adjusted pricing.

The article outlines 20 essential cybersecurity concepts, including phishing and zero-trust architecture, to help individuals protect their data and recognize online threats.