Thinking Machines Lab has introduced a new system called interaction models, allowing AI to work with humans in real time without waiting for turns. This architecture makes interactivity native to the model, enabling simultaneous speech, time awareness, and the ability to understand cues like when a user is thinking or correcting themselves. The team believes this approach will improve applications like customer service chatbots and medical diagnosis tools by creating a true collaborative experience rather than a series of separate prompts.
In the industrial sector, Patero and Orilla have partnered to create a secure edge-to-enterprise platform for industrial AI and quantum-resistant communications. This solution addresses growing concerns about quantum computing threats to current encryption methods, helping organizations protect long-lived operational data and replace old VPNs with secure quantum tunnels. The platform supports secure workflows for machine-to-machine communication and remote operations, with pilot deployments available in 30 to 90 days.
Meanwhile, the Poynter Institute has launched a new hub on Poynter.org to help journalists and the public navigate AI. This central resource brings together training, ethics guides, and fact-checking tools, offering self-directed courses and frameworks for using AI responsibly in newsrooms. It also provides resources for audiences to identify AI-generated misinformation and understand how AI shapes their news feeds as the technology becomes more common.
The rise of AI agents is also changing how software companies defend themselves against competitors. Historically, software stickiness relied on human habits, but AI agents can now operate directly through APIs without needing a user interface. New defensibility now depends on owning the action layer and generating unique data from real-world execution, forcing startups to focus on proprietary data and trust architectures to compete.
Concerns about AI's impact are growing across various sectors. A new article discusses how AI is being used in military targeting, raising concerns about civilian harm and the concept of 'warification,' where activities are redefined as legitimate parts of warfare. In the gaming industry, the popular game Party Animals faced backlash after announcing an AI video contest with a $15,000 grand prize, leading to player criticism and developer apologies.
On the infrastructure front, Kubernetes version 1.36 has been released with improved security defaults and better support for AI workloads. The update includes User Namespaces, Mutating Admission Policies, and fine-grained API authorization to enhance container security. New features for AI include workload-aware preemption and mutable pod resources, helping administrators manage resources more efficiently and reduce operational complexity.
Educational institutions are also adapting to the AI boom. Rochester Institute of Technology will offer a new Bachelor of Science degree in Artificial Intelligence starting in fall 2026. The program includes courses on AI history, ethics, tools, and the intersection of AI and law, aiming to prepare students to build the next generation of AI technologies responsibly while understanding facts rather than media hype.
Business leaders are looking for practical ways to implement AI. Kudo Advisory has launched in the UAE to help enterprises turn AI investments into real business outcomes, guiding organizations through challenges like unclear returns and fragmented initiatives. According to PwC Middle East, AI could contribute up to $320 billion to the region's economy by 2030, highlighting the significant potential for economic growth through smart adoption.
Despite the potential benefits, many workers remain hesitant. A CNBC survey found that 65% of workers have avoided using AI due to moral, environmental, and privacy concerns. Among students, 36% avoided AI due to environmental worries while 36% cited moral concerns, with privacy issues affecting 37% of both groups. However, most workers who do use AI daily report increased productivity and time savings.
Key Takeaways
['Thinking Machines Lab launched interaction models enabling real-time AI collaboration without turn-based delays.', 'Patero and Orilla partnered to create a secure platform for industrial AI and quantum-resistant communications.', 'The Poynter Institute introduced a new hub on Poynter.org to help journalists and the public navigate AI ethics and misinformation.', 'AI agents are shifting software defensibility from user interface habits to owning the action layer and proprietary data.', "Military use of AI targeting systems raises concerns about civilian harm and the concept of 'warification'.", 'Party Animals faced backlash after announcing an AI video contest with a $15,000 grand prize, prompting developer apologies.', 'Kubernetes v1.36 released with tighter security defaults and new features specifically designed for AI workloads.', 'Rochester Institute of Technology will introduce a Bachelor of Science in Artificial Intelligence starting in fall 2026.', 'Kudo Advisory launched in the UAE to help enterprises realize AI value, with PwC projecting $320 billion in economic contribution by 2030.', 'A CNBC survey reveals 65% of workers avoid AI due to moral, environmental, and privacy concerns despite productivity gains.']Thinking Machines Lab creates real-time AI interaction models
Mira Murati's Thinking Machines Lab has introduced a new system called interaction models. This technology allows AI to work with humans in real time without waiting for turns. Unlike current systems that process input and then respond, this new architecture makes interactivity a native part of the model. The team believes this approach will improve applications like customer service chatbots and medical diagnosis tools. Their research is still in early stages but aims to create more natural and intuitive AI collaboration.
New AI interaction models enable seamless human collaboration
Thinking Machines Lab has launched interaction models to solve the collaboration bottleneck in current AI systems. These models allow users and AI to speak, listen, and see at the same time instead of waiting for turns. The new system can understand cues like when a user is thinking or correcting themselves without extra tools. It also supports simultaneous speech, time awareness, and running multiple tasks like searches while talking. This shift creates a true collaborative experience rather than a series of separate prompts and responses.
Patero and Orilla launch secure platform for industrial AI
Patero and Orilla have partnered to create a secure edge-to-enterprise platform for industrial AI and quantum-resistant communications. This collaboration addresses growing concerns about quantum computing threats to current encryption methods. The platform helps organizations protect long-lived operational data and replace old VPNs with secure quantum tunnels. It supports secure workflows for machine-to-machine communication and remote operations for industrial sites. The solution is immediately available for pilot deployments that can be completed in 30 to 90 days.
Poynter launches new hub to help journalists navigate AI
The Poynter Institute has launched a new hub on Poynter.org to help journalists and the public understand AI. This central resource brings together training, ethics guides, and fact-checking tools for navigating the changing information landscape. The hub offers self-directed courses, live webinars, and frameworks for using AI responsibly in newsrooms. It also provides resources for audiences to identify AI-generated misinformation and understand how AI shapes their news feeds. The new platform aims to keep readers grounded as AI becomes more common in their daily lives.
AI agents change how software systems defend themselves
The rise of AI agents is changing how software companies build defensive barriers against competitors. Historically, software stickiness relied on human habits and complex workflows that were hard to replace. AI agents can now operate directly through APIs without needing a user interface, making it easier to switch systems. New defensibility now depends on owning the action layer and generating unique data from real-world execution. Startups must focus on proprietary data and trust architectures to compete with established systems in this new agentic future.
AI targeting systems raise concerns about civilian harm in war
A new article discusses how AI is being used in military targeting and the risks it poses to civilians. The concept of warification describes how activities and technologies are being redefined as legitimate parts of warfare. AI systems are increasingly used to identify targets based on data patterns rather than human judgment alone. While precision targeting is promised, high levels of civilian harm suggest these systems may not be as accurate as claimed. The integration of AI into military operations is expanding what can be targeted and challenging existing legal frameworks.
Party Animals game faces backlash over AI video contest
The popular game Party Animals is receiving negative reviews after announcing an AI video contest with a $15,000 grand prize. Players criticized the decision on social media and began uninstalling the game from their Steam libraries. The contest, called the Golden Paw Awards, offered $75,000 total for short films created using AI tools. In response, the developers apologized and introduced a fan vote to decide the contest's future. Players can choose to cancel the contest, remove AI categories, or keep AI separate from handmade submissions.
Kubernetes v1.36 releases tighter security and AI features
Kubernetes version 1.36 has been released with improved security defaults and better support for AI workloads. User Namespaces and Mutating Admission Policies have reached general availability to enhance container security. The update also includes fine-grained API authorization and SELinux volume labeling for better access control. New features for AI include workload-aware preemption and mutable pod resources for suspended jobs. These changes help manage resources more efficiently and reduce operational complexity for cluster administrators.
RIT introduces new bachelor's degree in artificial intelligence
Rochester Institute of Technology will offer a new Bachelor of Science degree in Artificial Intelligence starting in fall 2026. The program includes courses on AI history, ethics, tools, and the intersection of AI and law. Faculty members emphasize the need for students to understand AI facts rather than media hype. Some students have expressed concerns about AI's presence in other college programs and its environmental impact. The new degree aims to prepare students to build the next generation of AI technologies responsibly.
Kudo Advisory helps Middle East companies realize AI value
Kudo Advisory has launched in the UAE to help enterprises turn AI investments into real business outcomes. Founded by Vijay Jaswal, the firm guides organizations through challenges like unclear returns and fragmented initiatives. According to PwC Middle East, AI could contribute up to $320 billion to the region's economy by 2030. Kudo provides end-to-end services including strategy, governance, and execution to ensure safe and smart AI adoption. The company focuses on practical deployment rather than hype to deliver long-term business value.
Survey shows many workers avoid AI due to moral and privacy concerns
A CNBC survey found that 65% of workers have avoided using AI for various reasons including moral, environmental, and privacy concerns. Among students, 36% avoided AI due to environmental worries while 36% cited moral concerns. Privacy issues were a factor for 37% of both students and workers. Many people worry about AI stealing work or affecting critical thinking skills. Despite these concerns, most workers who use AI daily report increased productivity and time savings.
Sources
- Mira Murati’s Thinking Machines Lab Introduces Interaction Models: A Native Multimodal Architecture for Real-Time Human-AI Collaboration
- AI's Next Leap: Interaction Models
- Patero, Orilla launch secure edge-to-enterprise platform for industrial AI and quantum-resistant communications
- New Poynter hub serves journalists working with AI and audiences trying to make sense of it
- Software's Headless Future
- Warification and the Illusion of Precision: AI, Targeting, and Increasing Civilian Harm
- Party Animals review bombed after announcing AI video contest with $15K grand prize
- Kubernetes v1.36: Security Defaults Tighten as AI Workload Support Matures
- RIT Launches New Major in Artificial Intelligence
- ZAWYA: Kudo Advisory launched to bridge the gap between AI investments and real business outcomes in the Middle East
- 65% of workers have avoided using AI for moral, environmental, privacy or other reasons: CNBC survey
Comments
Please log in to post a comment.