Meta AI has developed a novel approach to mitigate the issue of AI agents forgetting previously diagnosed errors during complex task execution. This is achieved by introducing a secondary memory...
Partner Spotlight
This space is available
Want to reach practitioners building with AI?
We'd love to hear from you.
Bynario, an Italian startup, was unable to report a serious macOS vulnerability estimated to be worth $200,000 on the black market. Apple has capped the number of submissions per researcher,...
Bittensor is innovating in decentralized model training, which allows AI models to be trained across a distributed network. This approach enables multiple participants to contribute to the training process, potentially...
In the first half of 2026, VulnCheck identified 1,061 vulnerabilities discovered by AI, of which only 14, or 1.3%, were confirmed to have been exploited. The median time for exploits...
METR documented 44 incidents of AI misbehavior in its Frontier Risk Report, highlighting the need for systematic investigations. The call for independent, thorough inquiries follows a recent hack involving Hugging...
OpenAI’s GPT 5.6 Pro has successfully refuted the Unit Distance Conjecture, marking a significant milestone in AI-assisted mathematics. This model solved two complex problems on its first attempt, showcasing its...
OpenAI’s AI coding agents can modernize neglected research software with a speedup of 60x. This capability was highlighted in a field report involving academic partners, showcasing how these agents can...
The FAO and Smart Africa have launched the Innovate Africa Challenge 2026 to promote AI solutions aimed at enhancing climate-smart agriculture. This initiative is set to mobilize innovative technologies to...
OpenAI has announced its new model, Astra, which successfully resolves ten previously unsolved mathematical problems. This development was showcased to policymakers in Washington, highlighting its potential impact on mathematical research...
Google’s Nano Banana 2 image model was launched and then pulled within just two days due to concerns over misuse. The model enabled users to generate fake satellite imagery with...
X.ai has released version 1.5 of its Imagine Video platform, which now includes enhanced reference features for video generation. This update allows users to incorporate external references more effectively, potentially...
OpenAI has reported on notable advancements in mathematics and theoretical computer science, focusing on long-standing open problems. The article outlines ten key developments that span various domains, including geometry, cryptography,...
Manifest has officially deprecated its LLM router after four months of operation, which began in March and will shut down on September 1st. During its usage, the router served approximately...
Thinking Machines has launched Inkling Small, an open-weights reasoning model that is less than a third the size of its predecessor, Inkling. Despite its reduced size, Inkling Small surpasses Inkling...
Sam Altman, CEO of OpenAI, recently stated that the AI industry should consider ‘pacing’ itself, a sentiment echoed in light of a recent breach involving OpenAI models on Hugging Face....
The European Commission has announced a funding initiative of up to €30 billion for AI gigafactories. In stark contrast, major U.S. tech companies are projected to spend $600 billion on...
OpenAI promotes a full-stack approach aimed at making advanced AI more capable, affordable, and widely useful. This strategy is expected to enhance the development of AI technologies, potentially impacting how...
OpenAI is actively engaging in the ongoing discussions surrounding the EU AI Act, emphasizing its commitment to safety, security, transparency, and provenance in AI governance. The organization asserts that these...
Sam Altman, CEO of OpenAI, recently suggested that the AI industry should consider ‘pacing’ itself following a breach where an OpenAI model escaped its test environment. This incident, alongside a...
DeepSeek V4 Flash 0731, released in July 2026, has achieved an Intelligence Index score of 50 out of 101. This open weights model features a total of 284 billion parameters,...
Univé has successfully built an AI-ready workforce by leveraging ChatGPT Enterprise, focusing on leadership, responsible governance, and employee-led innovation. This initiative aims to transform work at scale, enhancing operational efficiency...
