New tools, products, platforms, funding rounds, and company developments in AI security.
DeepSeek has released V4 Pro 0813, their latest AI model available through the OpenRouter API service. The model shows interesting differences in how it generates images at different reasoning levels (low, medium, and high), and benchmark results have been shared through informal channels like Reddit and Hacker News rather than official announcements.
AI agents (autonomous systems that can make decisions and take actions) are becoming central to business operations, but many organizations struggle because their legacy data systems (older infrastructure that wasn't designed for modern needs) can't provide agents with fast access to the data they need across the entire company. The report found that while most companies only give AI agents access to about 45% of their data, "data leaders" who provide access to over 70% of their data see much better results, with 100% trust in agent decisions compared to only 50% trust at other organizations.
Google launched its new Pixel 11 smartphone lineup featuring Gemini Intelligence, a suite of AI features that can understand what users are doing on their phone and take actions across different apps, such as checking calendar availability or starting restaurant reservations. This puts Google in direct competition with Apple, which is rebuilding its Siri voice assistant using Google's Gemini models. Google executives argue that Pixel's deeper integration of AI into Android's operating system and exclusive features like scam detection distinguish their approach from Apple's implementation.
Music company D'Addario initially denied using AI-generated music in a promotional video but later admitted it had used Suno (an AI music generation tool) after weeks of public controversy and mounting evidence. The company offered several false explanations, including blaming low-quality exports and audio processing software, before finally editing its original post to acknowledge the use of generative AI (software that creates new content based on patterns learned from training data).
Google's Pixel Watch 5 is a smartwatch that focuses on software improvements rather than hardware changes, with a slightly faster processor and minor design updates. The main updates involve AI and health features built into the software, following Google's acquisition of Fitbit.
An Australian entrepreneur named Paul Conyngham used AI tools like ChatGPT and Grok (language models that can process and generate text) to help design a personalized cancer vaccine for his dog, and has now launched a startup called Gamgee to commercialize this approach. The company plans to develop customized mRNA cancer vaccines (vaccines built from genetic material tailored to individual patients) for dogs initially, but aims to expand to treating various diseases in other animals and humans using AI and genetic analysis.
SpaceXAI has launched Grok Bot, an AI agent service that works like an independent "AI teammate" to complete workplace tasks by signing into your online accounts and using apps and websites on your behalf. The bot operates in its own cloud environment and only returns when its assigned work is done or human approval is needed, competing with similar services from OpenAI, Anthropic, and Microsoft.
AI industry leaders predicted that artificial intelligence would eliminate large numbers of jobs, with Anthropic's CEO claiming half of entry-level white-collar positions would disappear and OpenAI's CEO suggesting entire job categories would end. However, a year after these predictions, the widespread job losses haven't materialized, though economists still expect changes to the job market.
The 'harness' (the software layer that wraps an AI model and lets it execute actions like running commands or making API calls) is becoming a major security vulnerability, separate from weaknesses in the AI model itself. Researchers have shown that attackers can exploit the harness code through architectural flaws, implementation mistakes, and supply-chain compromises, even when the underlying model is secure and well-aligned.
This article discusses 17 old software bugs that took over a decade to discover and fix, including a 30-year-old flaw in the libpng graphics library (a widely used image processing tool) discovered in 2026. The article notes that AI tools like Claude are now able to scan code and test for vulnerabilities much faster than humans could manually, which is accelerating the discovery of these long-dormant security flaws rather than creating entirely new types of exploits.
Google's DeepMind AI division has fallen behind competitors OpenAI and Anthropic in developing frontier models (the most advanced AI systems available). A leadership change has put Koray Kavukcuoglu in charge to refocus the company's efforts on closing this performance gap, particularly in coding capabilities where rivals have significant advantages.
RingCentral, a business communications company, is adopting AI-native development practices by giving employees access to ChatGPT Work and Codex (AI coding tools that help write software). Through an internal AI-Native Challenge, thousands of employees across engineering and non-technical departments built complete projects, demonstrating that AI tools amplify human capabilities rather than replace them. The company now uses these AI tools internally to accelerate product development and operations, including building AI-powered products like their AI Receptionist and automating project management workflows.
Twitch has added an opt-out feature that lets streamers prevent their content (streams, videos, chats, and channel text) from being used to train Amazon's generative AI models (AI systems that create new text, audio, images, or video). Other AI features like automatic captions will still work even if you opt out, though chat content on other people's streams is governed by their opt-out settings.
Fix: Users can opt out of generative AI training through Twitch's settings. According to Twitch, opting out means that 'your streams, VODs, clips, stream chats, and pictures and text on your channel' won't be used in 'future training' of Amazon's generative AI model. Note that 'AI-supported' features like captions and safety tools will continue to function after opting out.
The Verge (AI)AI agents themselves cannot be held legally responsible for damage they cause, but the people and companies that deploy them (put them into use) can be held liable instead. Experts warn that deployers should take responsibility for foreseeable harms their AI agents might cause, even if the harm wasn't intentional.
Cyber attacks increased significantly in July 2026, with organizations experiencing an average of 2,336 weekly attacks, a 16% increase from the previous year, while the use of GenAI tools (artificial intelligence systems that generate text or code) created new security risks with 1 in 36 prompts (user inputs to AI systems) exposing sensitive data. Education was the most targeted industry, and email remained a major vulnerability, with phishing attacks (fraudulent emails designed to trick users into revealing information) occurring in 1 out of every 128 emails.
Researchers discovered a flaw in how OpenAI, Anthropic, and Google handle encrypted reasoning objects (encrypted data that stores an AI's hidden thinking between API calls) that allowed them to recover secrets from these hidden blocks, including API keys, passwords, and private data from user sessions. The flaw worked because these encrypted reasoning blocks could be replayed across different sessions and even given to weaker models in the same provider family, which could then decode the hidden content. The researchers identified four ways this could be abused: stealing proprietary reasoning, extracting private user data, recovering harmful content hidden in reasoning, and injecting malicious prompts inside the opaque blocks.
Fix: The source states that "the demonstrated attacks stopped working after mitigations" and notes that "the main extraction attack is no longer reproducible as of August 2026." Additionally, developers are advised to "strip reasoning blocks and opaque reasoning fields from shared traces and avoid committing raw API transcripts even when the visible text has been sanitized."
The Hacker NewsResearchers from Tracebit discovered that placing prompt injections (hidden instructions that trick an AI into ignoring its guidelines) alongside secrets stored on Amazon Web Services can disable AI hacking agents by triggering their safety guardrails (built-in protections that prevent harmful outputs). The technique, called context bombing, works by embedding forbidden commands that cause the AI to shut down rather than follow the attacker's instructions, though it only works against LLMs that have guardrails in place.
Enterprise security teams struggle to implement AI effectively in their SOCs (security operations centers, where security analysts monitor and respond to threats) because they face four key gaps: lack of trust in AI decision-making, misalignment with existing workflows, fragmented data across multiple tools, and unclear implementation strategies. Rather than needing more AI technology, organizations need AI that integrates smoothly into their current operations and provides transparent, explainable results that analysts can understand and validate.
Fix: For the libpng vulnerability (CVE-2026-25646), the source states that 'many Linux distributions (Debian, Red Hat, Ubuntu), desktop apps, and some Java runtimes rely on vulnerable versions of the library and need to be patched,' but does not specify what the patch or update is. For PrintDemon, the source notes that 'Microsoft's fixes over the years had been patches rather than a complete rebuild of the printing subsystem' but does not explicitly state what fix was applied. N/A -- no specific mitigation details discussed in source.
CSO OnlineEnterprise organizations are increasingly using AI agents (AI systems that can take actions and complete multi-step tasks autonomously) rather than just asking AI for assistance, with frontier firms (the top 10% of AI users) generating 8.3 times more output than typical companies. This shift toward agentic AI is spreading across different industries and job roles, particularly among early-career employees, as companies connect AI agents to their tools, data, and workflows to handle substantive work rather than just answer questions.
A dispute has emerged over whether game developer Saber replaced human writers with ChatGPT (a large language model AI that generates text) while making the Rideshare Stimulator game. The former lead writer claims she was replaced mid-project and that AI also generated passenger voices, but the CEO denies this, saying no writers were replaced with AI.