Modelos, agentes, investigación y negocio de la IA.
Proton has launched Lumo 2.0, a massive update to its privacy-focused AI assistant. According to independent benchmarks, performance jumps up to 240% compared to the previous version, approaching ChatGPT and Claude. The big difference: Proton still does not train its models with user data, a European-style approach that could redefine trust in AI.
Vercel launches a public beta of an expanded version of its AI agent integrated into the dashboard, capable of answering questions, investigating production failures, and executing actions upon approval. The novelty lies in its contextual access to real deployment data and its granular control model.
AWS has increased Amazon Bedrock AgentCore execution quotas up to five times, allowing companies to support more concurrent agents and simultaneous sessions without requesting quota increases. The move responds to the rapid transition from pilot projects to production deployments of AI agents.
Large language models (LLMs) tend to give predictable, similar responses—a 'groupthink' problem. Australian startup Springboards has developed Flint, a model designed to offer greater variety and creativity in open-ended tasks like travel planning or brainstorming.
NASA, in collaboration with Red Hat, has developed an AI-powered medical assistant that works without an internet connection, designed for deep space missions. The system, called Crew Medical Officer Digital Assistant (CMO-DA), uses the open source tool RamaLama to run language and vision models locally on HPE hardware.
Anthropic has unveiled Claude Sonnet 5, a model designed to run AI agents more cost-effectively. With reduced prices and improved safety, it aims to democratize the use of autonomous agents in businesses.
The Bank of England warns that autonomous AI trading agents could trigger severe market disruptions if they react uniformly, creating dangerous feedback loops. Deputy Governor Sarah Breeden calls for new regulatory frameworks to address the systemic risk.
The dispute between Anthropic and the Pentagon is not just a contractual disagreement. Court emails show a clash of principles over whether a private company can impose barriers on the military use of its frontier AI, and what role the government should play in that decision.
Meta has imposed restrictions on its engineers to avoid using rival AI tools like Claude Code and Codex, aiming to prevent knowledge distillation and accelerate development of its own AI-assisted coding solutions.
Artificial intelligence is showing symptoms similar to the early days of cloud computing: the real challenge is no longer having the best model, but operating it reliably, efficiently, and securely at scale. The article analyzes the four key operational disciplines to face this new phase.
SAP has conducted its second executive reorganization of the year, dividing AI responsibilities between CEO Christian Klein and COO Sebastian Steinhäuser. The move reflects pressure to keep pace with competitors integrating AI into business processes more quickly.
Anthropic has partnered with the California government to provide its AI tools, including the Claude chatbot, to state agencies at a 50% discount. The agreement, announced Monday, aims to increase government efficiency by automating tasks such as information analysis and document drafting, without replacing human work.