AIが期間内の動向を整理
AI companies advance usability and infrastructure at the same time: the latest from OpenAI, Google, and Anthropic
The AI news on June 1 focused less on new features for their own sake and more on practical usability. OpenAI improved ChatGPT’s readability, Google expanded support for research and creative work, and Anthropic strengthened Claude’s infrastructure and European presence. For business users, output quality, ease of adoption, and operational stability are becoming even more important.
SOURCE CHECK
Primary Sources 18
Primary Sources
SOURCE CHECK 1
openai.com
SOURCE CHECK 2
deepmind.google
SOURCE CHECK 3
anthropic.com
SOURCE CHECK 4
x.com
SOURCE CHECK 5
x.com
SOURCE CHECK 6
x.com
SOURCE CHECK 7
forum.openai.com
SOURCE CHECK 8
openai.com
SOURCE CHECK 9
anthropic.com
SOURCE CHECK 10
anthropic.com
SOURCE CHECK 11
openai.com
SOURCE CHECK 12
openai.com
SOURCE CHECK 13
anthropic.com
SOURCE CHECK 14
openai.com
SOURCE CHECK 15
docs.anthropic.com
SOURCE CHECK 16
docs.anthropic.com
SOURCE CHECK 17
docs.anthropic.com
SOURCE CHECK 18
anthropic.com
Key Points
- 1OpenAI updated GPT-5.5 Instant to make ChatGPT responses easier to read
- 2Google DeepMind opened Co-Scientist to individual researchers for hypothesis generation
- 3Anthropic expanded compute partnerships with Google, Broadcom, and Amazon
- 4Anthropic opened a new Milan office to broaden European support
- 5OpenAI also shared guidance on prompt injection defenses and an audio API update
OpenAI: Making ChatGPT responses easier to read
OpenAI updated GPT-5.5 Instant in both ChatGPT and the API, improving readability and naturalness. The changes reduce overly long responses and bullet-heavy outputs, making the model better suited for everyday advice and drafting. For users, readability affects productivity just as much as the content itself.
Google: Expanding support for research and creative work
Google DeepMind’s Co-Scientist now offers Hypothesis Generation for individual researchers, helping with the early stages of research. Because multiple agents generate and refine hypotheses, it is well suited to organizing issues and checking for overlooked points. Google’s Flow Agent also supports planning, editing, and organizing creative work, making complex tasks easier to move forward.
Anthropic: Strengthening the infrastructure and footprint behind Claude
Anthropic expanded compute partnerships with Google, Broadcom, and Amazon to secure the infrastructure needed to train and deliver Claude. It also opened a new office in Milan to broaden support in Europe. For customer organizations, these moves provide context for evaluating adoption support and operational stability.
Safety: What to watch when AI handles external content
OpenAI outlined defenses against prompt injection, the problem of external text trying to manipulate an AI. The key point is not to rely on filters alone, but to design systems that limit the impact of attacks. If your organization uses AI to read internal documents or external materials, safety design should be considered from the start.
Related News Items
Anthropic / press_release
Anthropic expands compute partnership with Google and Broadcom
OpenAI / feature
ChatGPT updates GPT-5.5 Instant for clearer responses
Anthropic / availability
Anthropic opens a Milan office to expand in Europe
Anthropic / press_release
Anthropic deepens Amazon partnership for more compute
OpenAI / prompt_pattern
OpenAI explains how to defend against prompt injection
Anthropic / prompt_pattern
Anthropic explains how to improve complex prompts
Anthropic / press_release
Anthropic updates Project Glasswing with clearer collaboration scope