OpenAI is also changing how it plans to make AI-generated text traceable. On 5 October 2026, the company introduced textGrain, an invisible statistical watermark for text. The technology is due to roll out gradually over the coming weeks for eligible ChatGPT and Codex outputs in the European Union, across all subscription tiers. For API customers, textGrain is already available worldwide as an optional feature for selected models, although it remains switched off by default.

Since 7 October 2026, Google’s SynthID Detector has been publicly available worldwide. At synthid.com, users can upload images, videos and audio files to check whether they contain an invisible SynthID watermark. The interface is initially available only in English. Before this public launch, the portal had been available since 2025 mainly to journalists, media professionals and selected researchers.

With Mistral Large 4, French AI company Mistral AI unveiled its largest and most capable model to date on 6 October 2026. Internally, it carries the rather unapologetic nickname “Le Chonk”. But there is more behind the unusual name than sheer scale: with its latest model, Mistral is attempting to narrow the gap between European open-weight systems and the most capable models emerging from the United States and China.

Germany has reached a notable point in its AI strategy. After years dominated by research, model development and policy frameworks, the emphasis is increasingly shifting towards practical deployment. Artificial intelligence is now being tested under real-world conditions across local government, industry, robotics, health research and education. At the same time, regulatory sandboxes, sovereign cloud infrastructure and major European programmes for industrial AI are beginning to take shape. The decisive shift is therefore away from broad AI experimentation and towards concrete, measurable applications that can be integrated into existing processes on a permanent basis.

Just a few months ago, the direction of travel seemed obvious: more powerful models, larger data centres, faster releases and increasingly autonomous AI agents. Now the tone across the industry is beginning to change. Senior figures at Anthropic and OpenAI are openly discussing whether the development of the most capable AI systems should proceed more slowly. Other influential voices across the sector are beginning to support at least the basic principle.

With GPT-6 Astra, OpenAI has introduced its most capable AI model to date. Astra began rolling out on 3 September 2026 and is already available via the OpenAI API. Access through ChatGPT is being introduced gradually, meaning availability depends on your subscription tier, workspace configuration and rollout status.

When people talk about the race for artificial intelligence, they usually think of new models from OpenAI, Google, Anthropic or DeepSeek. In reality, however, the next phase of AI development is increasingly being decided elsewhere: in computing infrastructure. Germany is therefore investing billions of euros in new data centres. As of today, more large-scale AI data centre projects are being planned or built than ever before. The objective is clear: expand domestic computing capacity, strengthen digital sovereignty and reduce dependence on the major US cloud providers.

From tomorrow, 24 August 2026, OpenAI will officially launch ChatGPT Ads in Germany and 30 other European markets. For the first time, advertisements will appear directly within ChatGPT, although initially only for users on the Free and ChatGPT Go plans. At launch, OpenAI is deliberately introducing non-personalised advertising across the European Economic Area (EEA).