Hub Artificial Intelligence

Artificial intelligence is either a field of research, the most important technical innovation of mankind, its downfall – or simply a pipe dream from Silicon Valley. Can machines really be intelligent? What is intelligence anyway? What opportunities does AI technology offer, and what are the risks? From neural networks to the science fiction vision of super AI, from deepfakes to AI surveillance: THE DECODER delivers the latest AI news and information on all facets of artificial intelligence.

Artificial Intelligence

OpenAI launches new ChatGPT agent that automates complex tasks for Pro, Plus, and Team

Artificial Intelligence

OpenAI releases its first open-weight language models since GPT-2 with GPT-oss

Artificial Intelligence

Google upgrades Gemini with Deep Think and flags early warning risks

Ad
Artificial Intelligence
Short

OpenAI has appointed Fidji Simo as CEO of its new Applications division, reporting directly to OpenAI CEO Sam Altman. Simo will be responsible for driving growth in core business areas such as product and operations. Previously, she served as Chair of the Board at Instacart and was already a member of OpenAI's board of directors. In its own words, the move reflects OpenAI's efforts to expand its organizational structure as it moves from a pure research organization to one focused on building products and scaling infrastructure, along with nonprofit initiatives. Altman remains CEO of OpenAI and will focus more on research, compute, and safety. Simo plans to step down from her role at Instacart in the coming months and join OpenAI later this year.

Artificial Intelligence
Short

Anthropic has launched a web search feature for its Claude API, letting developers combine Claude models with up-to-date web data without building their own search infrastructure. Claude decides when a web search makes sense based on the user's prompt, generates targeted queries, analyzes the results, and delivers answers with source links. The system can also handle multi-step research tasks. Web search is available for Claude 3.7 Sonnet, 3.5 Sonnet, and 3.5 Haiku, and costs $10 per 1,000 searches. Organizations can allow or block specific domains and manage web search access at the organizational level. The feature is also available for Claude Code, which can use it to look up API documentation or technical articles.

Ad
Ad
Artificial Intelligence
Short

Microsoft's Phi 4 model generates 56 sentences before responding to "Hi", developer Simon Willison found. This behavior, known as "overthinking", was confirmed by Microsoft's Dimitris Papailiopoulos, who says it's problematic for simple tasks but intentional for complex ones. He plans to address the issue. Microsoft released the open Phi 4 reasoning models in early May.

Phi 4's reasoning process shown in screenshot by Simon Willison.
Ad
Ad
Ad
Ad
Artificial Intelligence
Short

IBM CEO Arvind Krishna says the company has used AI and so-called AI agents to cut several hundred jobs in its HR department. At the same time, IBM has created new positions in areas like software development, sales, and marketing—roles Krishna says still require human judgment, while routine tasks are increasingly automated. Despite the growing use of AI, Krishna says IBM's overall workforce has increased, as automation frees up resources for what he calls "critical" activities. IBM has also introduced new services that let companies build and manage their own AI agents. These tools are designed to work with solutions from other providers, including Amazon, Microsoft, and OpenAI. According to the company, IBM has already signed $6 billion worth of consulting contracts in the generative AI field.

WSJ
Google News

Bank details

IBAN: DE88 2507 0070 0053 0014 00
BIC: DEUTDE2HXXX
Account holder: Deep Content GmbH
Purpose: Support THE DECODER
AI in practice

New data from OpenAI and Anthropic show how people actually use ChatGPT and Claude

AI and society

Leading AI chatbots are now twice as likely to spread false information as last year, study finds

AI in practice

Anthropic confirms technical bugs after weeks of complaints about declining Claude code quality

Google News