Anthropic has created a method to defend AI models against “jailbreaks” — unauthorized workarounds to get an AI model to do things it was trained not to do, like providing instructions for building chemical weapons. Called Constitutional Classifiers, the system was 95 percent effective in identifying and preventing jailbreaks of Anthropic’s Claude 3.5 Sonnet in a test environment. In an effort to drum up real-world red-teaming, the company offered cash prizes of up to $15,000 to anyone who could jailbreak its Sonnet AI model. After some 3,000 hours of attempts by 185 participants, none claimed an award. Now the company is offering additional incentives. Continue reading Anthropic Will Award Cash for Jailbreaking AI Defense System

An internecine AI battle has erupted between Alibaba and DeepSeek. Days after DeepSeek dominated several news cycles with its affordable DeepSeek-R1 reasoning model and the multimodal Janus-Pro-7B, Alibaba released its latest LLM, Qwen 2.5-Max, available via API from Alibaba Cloud. As with DeepSeek, Alibaba is looking beyond its domestic borders, but the fact that a public-facing AI battle is heating up between Chinese companies indicates the People’s Republic isn’t going to quietly cede the AI race to the U.S. Alibaba claims Qwen 2.5-Max outperforms models from DeepSeek, Meta and OpenAI. Continue reading Alibaba Plans to Take On AI Competitors with Qwen2.5-Max

Jack Dorsey’s financial tech and media firm Block (formerly Square) has released a platform for building AI agents: Codename Goose. Previously available in beta, Goose is primarily designed to build agents for coding and software development, but Block built in many basic features that could be applied to general purpose pursuits. Because it is open source and offered under Apache License 2.0, the hope is that developers will apply it to varied use cases. A leading feature of Codename Goose is its flexibility. It can integrate a wide range of large language models, letting developers use it with their preferred model. Continue reading Codename Goose: Block Unveils Open-Source AI Agent Builder

Perplexity joins the list of AI companies launching agents, debuting the Perplexity Assistant for Android. The tool uses reasoning, search, browsers and apps to help mobile users with daily tasks. Concurrently, Perplexity — independently founded in 2022 as a conversational AI search engine — has launched an API called Sonar intended for enterprise and developers who want real-time intelligent search, taking on heavyweights like Google, OpenAI and Anthropic. While to date AI search has largely been limited to answers informed by training data, which freezes their knowledge in time, next-gen tools can pull from the Internet in real time. Continue reading Perplexity Bows Real-Time AI Search Tool, Android Assistant

OpenAI has launched Operator, a semi-autonomous AI agent that uses a proprietary web browser to execute tasks like planning a vacation using Tripadvisor or booking restaurant reservations through OpenTable. “It can look at a webpage and interact with it by typing, clicking and scrolling,” explains OpenAI. Operator is powered by a new model called Computer-Using Agent (CUA), and is available in research preview to ChatGPT Pro subscribers in the U.S. Combining GPT-4o’s computer vision capabilities with advanced reasoning, CUA is trained to interact with graphical user interfaces (GUIs) — parsing menus, clicking buttons and reading screen text. Continue reading OpenAI Operator Agent Available to ChatGPT Pro Subscribers

Google has introduced Gemini 2.0, the latest version of its multimodal AI model, signaling a shift toward what the company is calling “the agentic era.” The upgraded model promises not only to outperform previous iterations on standard benchmarks but also introduces more proactive, or agentic, functions. The company announced that “Project Astra,” its experimental assistant, would receive updates that allow it to use Google Search, Lens, and Maps, and that “Project Mariner,” a Chrome extension, would enable Gemini 2.0 to navigate a user’s web browser to complete tasks autonomously. Continue reading Google Releases Gemini 2.0 in Shift Toward Agentic Era of AI

After years of focusing on AI infrastructure, Amazon is plunging into the frontier model business with the Nova series. The new family of generative AI models includes the text-to-text model Amazon Nova Micro and Amazon Nova Lite for fast, mobile-friendly apps, and at the upper echelon the multimodal Amazon Nova Pro and Amazon Nova Premier for processing text, images and video. Amazon, which is heavy into production via Amazon Studios and MGM, is also launched two specialty models focused on “studio quality” output — Amazon Nova Canvas for images and Amazon Nova Reel for video. Continue reading Amazon Dives into Generative AI with Nova Foundation Models

Amazon Web Services is building a supercomputer in collaboration with Anthropic, the AI startup in which the e-commerce giant has an $8 billion minority stake. Hundreds of thousands of AWS’s flagship Trainium chips will be amassed in an “Ultracluster” that when it is completed in 2025 will be one of the largest supercomputers in the world for model training, Amazon says. The company announced the general availability of AWS Trainium2-powered Amazon Elastic Compute Cloud (EC2) virtual servers as well as Trn2 UltraServers designed to train and deploy AI models and teased next-generation Trainium3 chips. Continue reading AWS Building Trainium-Powered Supercomputer with Anthropic

Artificial voice startup Hume AI has had a busy Q4, introducing Voice Control, a no-code artificial speech interface that gives users control over 10 voice dimensions ranging from “assertiveness” to “buoyancy” and “nasality.” The company also debuted an interface that “creates emotionally intelligent voice interactions” with Anthropic’s foundation model Claude that has prompted one observer to ponder the possibility that keyboards will become a thing of the past when it comes to controlling computers. Both advances expand on Hume’s work with its own foundation model, Empathic Voice Interface 2 (EVI 2), which adds emotional timbre to AI voices. Continue reading Hume AI Introduces Voice Control and Claude Interoperability

Anthropic is releasing what it hopes will be a new standard in data integration for AI. Called the Model Context Protocol (MCP), its goal is to eliminate the need to customize each integration by having code written each time a company’s data is connected to a model. The open-source MCP tool could become a universal way to link data sources to AI. The aim is to have models querying databases directly. MCP is “a new standard for connecting AI assistants to the systems where data lives, including content repositories, business tools, and development environments,” according to Anthropic. Continue reading Anthropic Protocol Intends to Standardize AI Data Integration

Amazon reported major revenue and profit increases during its third quarter, beating Wall Street’s forecasts, based largely on the company’s e-commerce sales and increasing demand for its cloud services. Capital expenditure, which reached a record amount following Amazon’s recent investments in artificial intelligence, will maintain its momentum as the company plans $75 billion capex on developing generative AI services over 2024-2025. “The faster we grow demand, the faster we have to invest capital in data centers, network gear and hardware,” explained CEO Andy Jassy. “We invest in all that upfront in advance of when we can monetize it.” Continue reading Amazon Pushes AI, Records Growth in Q3 Revenue and Profit

Microsoft is previewing GitHub Copilot for Azure in an ambitious expansion of its AI app development toolkit that some say could fundamentally change how developers build software for the AI era. The new premise is that switching from one software to another, as developers often do, should be seamless, not disruptive — sort of a real-time language translation and integration system for code. To fend off the move by Microsoft, AWS announced it is making its Q Developer AI code assistant available as an inline chat add-on accessible from IDEs like JetBrains and Microsoft’s own Visual Studio. Continue reading Microsoft, Amazon Jockey for Lead Among AI Code Assistants

In its first week of public beta, Anthropic’s “Computer Use” feature is gaining immediate traction, helping people do research and complete coding tasks. Claude works autonomously in Computer Use mode, suggesting broad implications for future productivity and workforce goals. Coming on the heels of OpenAI’s Swarm framework, these early forays into independent AI assistants seem to indicate that implementing such systems will be an area of focus for businesses in 2025. Claude can “see” what’s onscreen and use its “judgment” to adapt to different tasks, segueing across workflows and software. Continue reading Anthropic’s AI Agents for Claude Sonnet Increase Productivity

Penguin Random House, the world’s largest commercial book publisher, has updated the copyright disclaimer that appears in every book to say “no part of this book may be used or reproduced in any manner for the purpose of training artificial intelligence technologies or systems.” The warning will roll out globally on all new releases as well as backlist titles that are reprinted. Tom Weldon, CEO of Penguin Random House UK, has told staff the company will at its discretion “use generative AI tools selectively and responsibly, where we see a clear case that they can advance our goals.” Continue reading Penguin Random House Warns All Against AI Model Training

Nvidia has debuted a new AI model, Llama-3.1-Nemotron-70B-Instruct, that it claims is outperforming competitors GPT-4o from OpenAI and Anthropic’s Claude 3.5 Sonnet. The impressive showing has prompted speculation of an AI shakeup and a significant shift in Nividia’s AI strategy, which has thus far been focused primarily on chipmaking. The model was quietly released on Hugging Face, and Nvidia says as of October 1 it ranked first on three top automatic alignment benchmarks, “edging out strong frontier models” and vaulting Nvidia to the forefront of the LLM field in areas like comprehension, context and generation. Continue reading Nvidia’s Impressive AI Model Could Compete with Top Brands