AI Discovers Stronger, More Durable Plastics in Materials Science Breakthrough
Researchers at MIT and Duke University have utilized machine learning to develop significantly more durable plastics, a breakthrough that could lead to longer-lasting products and a reduction in plastic waste. The AI model identified novel crosslinker molecules, known as mechanophores, that can be integrated into polymer materials to make them more resistant to tearing. These mechanophores respond to mechanical force, allowing the material to withstand greater stress before breaking. After the AI model performed a rapid computational analysis of thousands of potential molecules, a task that would have been incredibly time-consuming with traditional methods, researchers synthesized a polymer with one of the identified candidates. The resulting material was found to be approximately four times tougher than plastics made with standard crosslinkers. This AI-driven approach to polymer science not only accelerated the discovery process but also uncovered new design principles for creating more resilient polymers.
Sources:
UK’s Alan Turing Institute Faces Whistleblower Complaint Over Governance and Culture
Staff at the Alan Turing Institute (ATI), the UK’s leading AI research body, have filed a whistleblowing complaint with the Charity Commission, citing concerns about the organization’s governance and internal culture. The complaint alleges a failure of the board of trustees to provide strategic direction and ensure accountability. It also describes an internal culture of ‘fear, exclusion, and defensiveness’ and a lack of oversight regarding senior leadership changes. This comes as the institute faces pressure from the government to revise its strategic focus, a key component of the UK’s AI strategy, and after roughly 10% of its workforce was notified of potential redundancy. The whistleblowers warn that the institute is at risk of collapse due to government threats over its funding.
Sources:
Dr. Karl Kruszelnicki Develops AI Chatbot to Combat Climate Change Misinformation
Renowned Australian science communicator Dr. Karl Kruszelnicki is developing an AI-powered chatbot to combat climate change skepticism. The ‘Digital Dr. Karl’ is being created to provide evidence-based answers to questions about the climate crisis, backed by reliable sources. The chatbot is built on an open-source large language model (LLM) from French company Mistral and has been trained on a vast collection of Kruszelnicki’s research, including academic papers and articles from reputable publications. The initiative aims to make credible information on climate science more accessible to a wider audience, as Dr. Karl himself receives hundreds of questions daily. The project acknowledges the ethical considerations surrounding AI but emphasizes the urgency of addressing climate change.
Sources:
Human vs. AI: Moderators More Accurate for Brand Safety, Study Reveals
A recent study by AI brand protection company Zefr has found that human content moderators are more accurate than AI models at identifying policy-violating material for brand safety purposes, although at a significantly higher cost. The research, which evaluated several multimodal large language models (LLMs), concluded that while AI is a much cheaper alternative, human reviewers remain superior, especially in classifying complex or nuanced content. The paper highlights the trade-off between cost and effectiveness that marketers face when trying to prevent their ads from appearing alongside inappropriate content. The findings suggest that a hybrid approach, combining the efficiency of AI with the contextual understanding of humans, is likely the most effective and economical strategy for digital content moderation in the context of brand safety.
Sources:
ACHILLES Project Launched to Simplify EU AI Act Compliance
A new Horizon Europe project called ACHILLES is working to help organizations navigate the complexities of the European Union’s AI Act. The project aims to provide a framework and an Integrated Development Environment (IDE) to facilitate the creation of trustworthy AI products that are sustainable, transparent, and secure. With the AI Act’s regulations beginning to take effect, ACHILLES seeks to bridge the compliance gap for innovators without hindering performance. The project is being validated through four real-world pilots in sectors including healthcare, security, and quality management. By integrating legal and ethical requirements directly into the development process, the ACHILLES IDE intends to make it easier for businesses to build AI systems that align with key principles of AI governance and compliance.
Sources:
Google Boosts AI Tools: Generative Video in Photos & Advanced NotebookLM Features
In recent updates, Google has integrated new generative AI capabilities across its services. Google Photos now features a generative AI tool that can transform photos into eight-second video clips with sound, a feature first introduced in Gemini. This is part of a broader initiative to centralize creative tools within a new “Create” tab in the Photos app, which started rolling out in the U.S. in August 2025. Additionally, Google’s AI-powered research tool, NotebookLM, has been updated to generate narrated video slideshows from research documents, complete with AI-generated visuals and quotes to make complex information more digestible. Google DeepMind also unveiled Aeneas, a new AI model designed to help historians interpret, restore, and attribute fragmented ancient texts, initially trained on Latin inscriptions.
Sources:
New Generative AI Tools for Creators: Hedra Live Avatars & GitHub Spark
A suite of new generative AI tools for creators and developers launched in late July and early August 2025. On July 22, Hedra introduced Live Avatars, a platform that generates lifelike, interactive video avatars from a single image with low latency. Following this, on July 23, GitHub Spark was launched in public preview, enabling users to create and deploy full-stack web applications using natural language processing (NLP) prompts. Runway also introduced Aleph on July 25, a model that allows for advanced real-time AI video generation and editing, such as removing objects and changing camera angles, through simple text commands. These launches highlight a trend towards more accessible and powerful AI-driven content creation and development platforms.
Sources:
Fortinet Expands FortiCloud with New Identity and Secure Storage Services
Fortinet has expanded its FortiCloud platform by introducing three new services: FortiIdentity, FortiDrive, and FortiConnect. These additions are designed to provide enhanced identity management, secure storage, and secure communication for enterprises, enhancing hybrid work security. The new services are integrated into the Fortinet Security Fabric, which allows for centralized visibility and consistent policy enforcement across various elements including users, devices, and applications. This expansion is part of Fortinet’s strategy to offer a unified platform that helps organizations consolidate their cloud security tools and streamline operations. The services are available through the FortiCloud portal and are also accessible on the marketplaces of AWS, Azure, and Google Cloud.
Sources:
Headlamp AI Assistant Plugin Released to Simplify Kubernetes Management
A new AI-powered plugin for the Headlamp Kubernetes web UI, named Headlamp AI Assistant, has been released to simplify Kubernetes management and troubleshooting. This new tool aims to assist users, from beginners to experts, in understanding and operating their Kubernetes (K8s) clusters more effectively. The announcement was made on the Headlamp blog on August 7, 2025.
Sources: