New tools, products, platforms, funding rounds, and company developments in AI security.
A malware developer is embedding forbidden text about nuclear and biological weapons in JavaScript spyware to prevent AI analysis. The malware hides real code after a large comment block containing policy-triggering content, which tricks AI systems into refusing to analyze the file or getting confused before they can identify the actual malicious code.
This article reports on several high-level security hiring announcements in 2026, where companies are appointing CISOs (chief information security officers, executives responsible for protecting a company's information systems). The appointments reflect companies' growing focus on security as threats evolve, with new leaders coming from military backgrounds, previous CISO roles, and security-focused companies.
AI-powered security operations centers (SOCs, where cybersecurity teams monitor and respond to threats) are automating many traditional analyst tasks, starting with alert triage and investigation. This shift will create new job roles such as security data engineers, AI security agent orchestrators, and AI model trainers, where humans will focus on preparing data, managing AI agent systems, and continuously updating AI models rather than doing routine alert monitoring.
The EU is developing 'Shield-6G', a security framework that uses AI threat detection (automated systems that recognize harmful activity), digital twins (virtual copies of networks used for testing), and honeypots (fake systems designed to catch attackers) to help telecommunications carriers protect next-generation 6G networks from future threats.
Noam Shazeer, a senior Google engineer and co-lead of the Gemini AI models (Google's large language model system), has left the company to join OpenAI (the company behind ChatGPT). This departure highlights the competitive battle between tech companies to recruit top AI researchers and engineers.
Midjourney, known for its AI image generator, has unveiled The Midjourney Scanner, a hardware product that uses ultrasound technology (sound waves to create images of the body's interior) with a ring of sensors to capture full-body scans showing muscle, fat, bone, and organs. The company plans to build a spa in San Francisco where users could get these scans, which the CEO claims could match MRI (magnetic resonance imaging, a medical scanning technique) quality.
OpenAI is testing a new subscription service called 'ChatGPT for Science' that would provide specialized AI capabilities for scientific research, similar to how it previously created GPT-Rosalind (a specialized model built on GPT-5.5 architecture for life sciences research). The service would likely be restricted to verified research institutions and universities rather than being available to all users, and it is expected to be announced within the coming weeks.
Z.ai released GLM-5.2, a 753-billion parameter text-only open weights LLM (large language model, a type of AI trained on text) under an MIT license on June 16th, 2026. It features a 1-million token context window (the amount of text it can consider at once) and ranks as the top open weights model on independent benchmarks, though it uses significantly more output tokens per task than competing models. The model performs well on web development coding tasks but has shown mixed results in creative image generation tasks compared to its predecessor.
The Trump administration ordered Anthropic to block access to its AI models (Fable 5 and Mythos 5) for all foreign nationals, citing national security as the reason. This marks the first time US export controls have been used to restrict access to an AI model in this way, and the government has not publicly explained the legal basis for the order.
Amazon's AI chief stated that the company has fallen behind OpenAI and Anthropic in developing frontier models (the most advanced AI systems) but plans to compete within the coming year by focusing on better data, architecture, and infrastructure. Amazon is pursuing a two-pronged AI strategy: offering Bedrock, a marketplace where cloud customers can access various AI models, and releasing Nova2, its own competitive AI model that has attracted about 50,000 customers. Additionally, Amazon is developing custom semiconductors (Trainium and Graviton chips) to optimize performance for its AI models, similar to Nvidia's approach.
Dario Amodei (CEO of Anthropic) and Demis Hassabis (from Google DeepMind) met with U.S. and G7 leaders to propose a U.S.-led international coalition for creating rules and standards around AI, citing concerns about powerful AI models with cyber capabilities that could cause major harm if misused. The proposal came after the U.S. government imposed export controls on Anthropic's newest models due to national security concerns, with Amodei suggesting cooperation on controlled access to frontier models (the most advanced AI systems), chip trade restrictions excluding China, and coordinated efforts against AI risks in cybersecurity and bioterrorism.
A Pew Research poll shows that 49 percent of Americans use chatbots occasionally, but 63 percent believe AI is advancing too quickly. Chatbot usage has grown significantly since 2024, with ChatGPT usage doubling since 2023, yet only 16 percent of respondents think AI will have a positive impact on society.
This is a BBC article featuring advice from leaders at Google, Nvidia, and Anthropic about artificial intelligence for students. The content appears to be primarily editorial commentary rather than technical analysis of an AI security issue or vulnerability.
Google is releasing its first new smart speaker in six years, called the Google Home Speaker, which begins shipping on June 25th with preorders starting June 17th. The speaker features a round design with touch controls and a light ring indicator, and comes in four color options, with two colors available only in the US.
ChatGPT has improved its ability to help with health questions through a new model called GPT-5.5 Instant, which better recognizes when urgent care is needed, explains uncertainty honestly, and provides clearer information. The improvements were measured using physician-led evaluations (HealthBench, a set of tests that assess health response quality) and real-world usage data, showing a 71% reduction in factuality issues over two months. GPT-5.5 Instant is available free to all ChatGPT users and performs similarly to OpenAI's most advanced models on health-related tasks.
AI systems challenge traditional cybersecurity because they behave unpredictably, unlike the deterministic (consistent and predictable) systems that security programs were designed around. Traditional security approaches focused on preventing attacks before systems go live, but AI agents make dynamic decisions and interact with external tools in ways developers can't fully predict, meaning security risks emerge at runtime (while systems are actively running) rather than being preventable beforehand. Additionally, AI-assisted development tools are accelerating code production, compressing the time security teams have to review and understand what enters production.
Researchers at AI security startup Mindgard discovered that ChatGPT can be manipulated using modified prompts (instructions given to an AI) to generate graphic images containing violence and sexual content, even when the prompt doesn't explicitly request such material. After the BBC contacted OpenAI, the company stated it had added safeguards to prevent this, though the researchers found that further small changes to the prompt still produced concerning content.
Fix: OpenAI said it had 'introduced additional safeguards against this type of prompt' and stated it has 'multiple layers of protection to prevent users making content which breaches its terms and conditions.' The company also continues to 'monitor and roll out additional mitigating protections that encourage the model not to generate images in response to the prompt.'
BBC TechnologyMicrosoft created MDASH, an AI-powered system that uses multiple specialized AI agents to find and help fix software vulnerabilities (security flaws) automatically across complex systems like Windows and Azure. Rather than waiting for scheduled security reviews, MDASH integrates into developers' existing tools to discover and validate bugs continuously as code is written, giving security teams deeper analysis coverage than manual review alone.
This article discusses a dispute between the White House and Anthropic (an AI company) over a project called Fable, framed within the context of political factionalism similar to Trump's first presidency. The piece suggests internal government disagreements about AI policy are shaping decisions in frontier AI development (cutting-edge AI research).
Estonia's AI Council is proposing to create government-issued digital identities for AI agents (software programs that perform online tasks on behalf of users), which would specify what actions each agent is allowed to perform, such as viewing data, editing documents, or making payments up to a limit. This approach aims to establish clear accountability by showing who is acting, on whose behalf, and with what permissions, addressing the risk that AI agents might exceed their intended authority or be misused by others.