Bookmarks (552)

AI agents will bring enterprises to the next level, but the same applies to related vulnerabilities. Here are key tips to follow.

Chains of smaller, specialized AI agents aren't just more efficient — they will help solve problems in ways we never imagined.

Sam Altman, CEO of OpenAI, admits the company was "on the wrong side of history" in the open source AI debate during a Reddit AMA, signaling a potential strategy shift as Chinese competitor DeepSeek disrupts the market with efficient, open models.

Nvidia confirmed that its CEO Jensen Huang met with U.S. President Donald Trump to discuss U.S. technology and AI leadership.

o1 is slightly better at reasoning, but DeepSeek-R1 provides much more details about its reasoning, which is very useful to the user.

It's 63% cheaper than OpenAI o1-mini and 93% cheaper than the full o1 model, priced at $1.10/$4.40 per million tokens in/out.

Last year, the saying going around the game industry was Survive until 25. The hope was that things would get better after a very rough year.

Palona’s approach suggests that customer relationships don’t have to be sacrificed for automation — instead, AI can enhance personalization.

French AI startup Mistral unveils a breakthrough 24B parameter language model that matches the performance of models three times its size, challenging tech giants with faster speeds and lower computing costs while targeting enterprise deployments.

Cerebras Systems launches DeepSeek's R1-70B AI model on its wafer-scale processor, delivering 57x faster speeds than GPU solutions and challenging Nvidia's AI chip dominance with U.S.-based inference processing.

Riffusion launches a free AI music generation platform that creates original songs from text and audio prompts, challenging tech giants with personalized learning features and backing from The Chainsmokers.

Y Combinator-backed startup Martin AI secures $2M seed funding to challenge Siri and Google with its innovative personal AI assistant, built by 19-year-old founders to revolutionize how consumers interact with AI through custom memory architecture and multi-channel accessibility.

DeepSeek-R1 released model code and pre-trained weights but not training data. Ai2 is taking a different approach to be more open.

Little Umbrella, an emerging pioneer in AI-powered social games, has raised $2 million in funding after the success of its Death by AI social game.

DeepSeek and its R1 model aren't wasting any time rewriting the rules of cybersecurity AI in real-time. Enterprises can't ignore this risk.

How PigAPI enables AI agents to interact directly with graphical user interfaces (GUIs) within virtual Windows desktops hosted in the cloud.

AI filmmakers and creatives applauded the move — welcoming the opportunity to receive copyright protections on their work.

Agents alone aren't enough for enterprise agentic AI workflows, that's where the power of AI orchestration could help to make a big difference.

The reduction in noisy, unstructured telemetry data by Observo can cut enterprise observability costs by up to 50%.

Existing open-source AI approaches are still not entirely open, which is a challenge that former Google and Apple engineers alongside a coalition of 13 universities are looking to solve.

AI agents are set to transform enterprise. But when will they really come to the fore? Salesforce and Mulesoft says this will be the year.

Alibaba's Qwen2.5-Max AI model sets new performance benchmarks in enterprise-ready artificial intelligence, promising reduced infrastructure costs and improved efficiency for business applications.

Endpoint, identity, and multi-domain attacks are dominating the enterprise threatscape today, fueled by new tradecraft invented using gen AI.

VB LAB INSIGHTS: This is a VB Lab Insights article presented by Capital One. AI offers transformative potential, but unlocking its value requires strong data management. AI builds on a solid data foundation that can iteratively improve, creating a flywheel effect between data and AI. This flywheel enables companies to build more customized, real-time solutions that unlock impact […]

Incention is launching a platform to bring AI and blockchain to Hollywood, and it is working with Hollywood writer David S. Goyer.

AI labs might be in panic mode, but enterprises are here to reap the rewards of the effects caused by DeepSeek R1.

DeepSeek's Janus Pro 7B vision AI model launch sends shockwaves through markets, triggering US tech stock sell-off and intensifying competition with Silicon Valley giants like OpenAI and Nvidia.

DeepSeek-R1 can collect and store on servers located in China, but there's an easy way to circumvent this.

For now, ChatGPT remains the better-rounded and more capable product, offering a suite of features that DeepSeek simply cannot match.

Haptikos has unveiled a hand exoskeleton, meant to enable you to use your hands with precision in haptic applications in XR.

Here's what OpenAI's o1 had to say about AI trends this year. Just how many there are is a testament to AI’s general-purpose value. 

DeepSeek R1’s Monday release has sent shockwaves through the AI community, disrupting assumptions about what’s required to achieve cutting-edge AI performance. This story focuses on exactly how DeepSeek managed this feat, and what it means for the vast number of users of AI models. For enterprises developing AI-driven solutions, DeepSeek’s breakthrough challenges assumptions of OpenAI's dominance -- and offers a blueprint for cost-efficient innovation.

Prosocial AI supports a hybrid future where human and machine intelligences co-evolve, guided by shared principles and grounded in truth.

Season 4 of a TV show is not something I write about very often, given my focus is video games. But Season 4 of Mythic Quest is another matter.

Marc Andreessen, Yann LeCunn and Mark Zuckerberg have all penned what appear to be responses to the Chinese open source model's ascent.

Y Combinator's fastest-growing startup Origami Agents raised $2M to build AI research tools that help sales teams find perfect customers.

OpenAI used its own o1-preview and o1-mini models to test whether additional inference time compute protected against various attacks.

DeepSeek has a free website and mobile app even for U.S. users with an R1-powered chatbot interface similar to OpenAI's ChatGPT.

Trump unveils ambitious $500 billion 'Project Stargate' AI initiative with tech giants OpenAI, Oracle, and SoftBank, sparking controversy as Elon Musk questions funding and environmentalists raise concerns about massive data centers.

Sakana found that self-adaptive models can modify their weights during inference to adjust behavior to new and unseen tasks.

The developers of games like Street Fighter are using generative AI to take the busywork out of development.

Hugging Face's new SmolVLM models run on smartphones, outperform larger systems and slash computing costs by 300X.

Instead of relying on specialized APIs, the system uses screenshots for visual input and virtual mouse and keyboard actions to complete tasks.

Pipeshift has a Lego-like system that allows teams to configure the right inference stack for their AI workloads, without extensive engineering.

Galileo launches Agentic Evaluations to help enterprises evaluate and monitor AI agents, securing $68M in funding as companies like Cisco adopt its platform for safer AI deployment.

Google challenges OpenAI with free Gemini 2.0 Flash Thinking model, offering million-token processing, native code execution, and breakthrough performance in math and science benchmarks.