The Builder's Brief
AI News: Extremism, Seismic Models, Acquisitions
Tuesday, April 28, 2026
Anthropic and OpenAI are exploring ways to combat extremism, with a crisis contractor in place. This move aims to address concerns about AI's potential to spread harmful content. The development is significant, as it shows AI companies taking proactive steps to mitigate risks. With the rise of AI, companies are under increasing pressure to ensure their technology is used responsibly.
Amazon Web Services (AWS) has announced the ability to scale seismic foundation models using Amazon SageMaker HyperPod. This development enables researchers to train larger models and expand context windows, leading to more accurate results. The advancement is crucial for the oil and gas industry, as it can help improve the accuracy of seismic data analysis. With this technology, companies can make more informed decisions about exploration and production.
Anthropic has acquired Vercept, a company that specializes in AI-powered solutions. The acquisition is expected to enhance Anthropic's capabilities in developing and deploying AI models. With this move, Anthropic aims to expand its offerings and improve its position in the market. The acquisition is a significant development in the AI industry, as it shows companies are investing in AI technology to drive growth.
The AI data centre boom is at risk of becoming a $9 trillion bust, according to a report. The rapid growth of AI has led to increased demand for data centres, but concerns about energy consumption and environmental impact are rising. As the industry continues to evolve, it's essential to address these challenges and ensure sustainable growth. The report highlights the need for more efficient data centres and renewable energy sources.
Researchers have developed a new benchmark, KidGym, to evaluate the abilities of Multimodal Large Language Models (MLLMs). The benchmark is inspired by the Wechsler Intelligence Scales and aims to assess MLLMs' capabilities in a more comprehensive way. The development of KidGym is significant, as it can help improve the performance of MLLMs in various tasks. By using this benchmark, researchers can identify areas where MLLMs need improvement and develop more effective training methods.
Researchers have introduced MemFactory, a unified framework for inference and training of agent memory in Large Language Models (LLMs). The framework aims to streamline the integration, training, and evaluation of memory-augmented LLMs. MemFactory is expected to facilitate the development of more capable and long-term AI agents. By providing a unified framework, researchers can focus on improving the performance of LLMs and developing more advanced AI models.
Baton is a new desktop app designed to simplify the development process with AI agents. The app allows users to manage multiple agents, worktrees, and projects in a single interface. Baton aims to improve productivity and streamline the development workflow. By providing a user-friendly interface, developers can focus on building and training AI models more efficiently. The app is expected to become a valuable tool for AI enthusiasts and professionals.
• Anthropic acquires Vercept
• Baton: AI agent desktop app
• AI data centre boom at risk
Think your CV is perfect? Let AI prove you wrong in seconds — brutal, honest, and hilarious feedback.
Try it free →🧠 Fun Fact: 45% of AI models use multimodal data
