Modelos, agentes, investigación y negocio de la IA.
GLM-5.2, Zhipu AI's new model, has demonstrated in independent tests cybersecurity performance comparable to Anthropic's Claude Mythos. Its open-source nature and free availability pose a geopolitical shift in AI.
Google has launched Nano Banana 2 Lite for image generation and Gemini Omni Flash for video, with unprecedented prices and latencies. This move democratizes visual content creation but also intensifies competition in the generative AI market.
Anthropic has identified in its Claude model an internal mechanism, called J-space, that allows representing concepts without needing to express them in the final response. The finding, not programmed but emerged during training, suggests that language models can develop a form of 'conscious access' similar to humans, with profound implications for AI interpretability and safety.
Recent advances in generative AI show the ability to write high-performance GPU kernels and automate complex freelance tasks. Meanwhile, analog computing promises extreme energy efficiency. Are we facing a new technological paradigm?
Nvidia's Kyber NVL144 rack, designed for Rubin Ultra GPUs, won't arrive until 2028 due to manufacturing challenges in its PCB backplane. The company has also canceled the NVL72x2 plan B due to customer rejection, leaving no scaling option for Rubin Ultra in 2027.
The AI revolution in healthcare promises efficiency, but it may dehumanize care. An analysis by TheVortiq reveals the risks of prioritizing technology over the patient.
Reddit has begun using language models (LLMs) to identify and remove spam generated by artificial intelligence. This strategy, reported by TechCrunch, reflects how platforms are forced to fight fire with fire in the era of generative AI.
Samsung Electronics reported an 1,800% increase in its second-quarter 2024 operating profit, driven by growing demand for artificial intelligence chips. This phenomenon reflects a structural shift in the semiconductor industry, where AI becomes the main growth driver.
For the first time, an AI agent carried out the technical execution of a real ransomware attack. However, a detailed analysis reveals that human intervention was indispensable: it selected the target, configured the infrastructure, and provided stolen credentials. This debunks the narrative of a fully autonomous debut in cybercrime.
A new wave of small artificial intelligence models is transforming sectors like healthcare, agriculture, and public health in regions with limited infrastructure. These systems work offline, consume little power, and run on low-cost devices, making them a viable alternative to large cloud-based models.
Researchers at Anthropic have discovered that Claude, their language model, has developed an internal structure called 'J-space' that functionally replicates the global workspace theory of consciousness. This finding, published in a study with 16 authors, could transform how we monitor the safety and transparency of AI systems.
AI browsers arrived to revolutionize web navigation, but a University of Washington analysis shows that four out of seven popular agentic browsers have critical vulnerabilities, including prompt injection attacks that can compromise user data. The article examines why these tools have failed to deliver on their promises and what risks they entail.