AI Act, AI Governance and AI Literacy in practice - analysis and commentary from the AI TrustCERT team.https://aitrustcert.com/en-UShttps://aitrustcert.com/en/news/ai-act-from-august-2-2026/https://aitrustcert.com/en/news/ai-act-from-august-2-2026/Five days before the AI Act took effect, the EU postponed obligations for high-risk systems to 2027 and 2028. What remains in force today - the Article 50 transparency obligations - affects a wider group of companies than it seems. Deadlines, fines, the Polish Act and a checklist for this week.Sun, 02 Aug 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-system-register-in-a-company/https://aitrustcert.com/en/news/ai-system-register-in-a-company/An AI system register is the foundation of EU AI Act compliance: without an inventory of tools, no obligation can be met. We explain what an AI tool register must contain, who should maintain it and how to build it step by step - with a practical field template.Fri, 03 Jul 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-act-fines-and-penalties/https://aitrustcert.com/en/news/ai-act-fines-and-penalties/What are the penalties for violating the EU AI Act? Up to EUR 35 million or 7% of turnover for prohibited practices, up to EUR 15 million or 3% for breaching obligations. We explain the penalty tiers, who they apply to and how to prepare before 2 August 2026.Fri, 03 Jul 2026 00:00:00 GMThttps://aitrustcert.com/en/news/company-ai-policy-checklist/https://aitrustcert.com/en/news/company-ai-policy-checklist/An AI usage policy tells employees which AI tools they may use, on what data and under whose oversight. A 10-point checklist of what practice and the EU AI Act require - plus the most common implementation mistakes.Fri, 03 Jul 2026 00:00:00 GMThttps://aitrustcert.com/en/news/eu-ai-act-compliance-audit-step-by-step/https://aitrustcert.com/en/news/eu-ai-act-compliance-audit-step-by-step/How to run an EU AI Act compliance audit? A practical guide: inventory, prohibited practices, risk classification, AI literacy, transparency and evidence. Check your organisation before the supervisory authority does.Fri, 03 Jul 2026 00:00:00 GMThttps://aitrustcert.com/en/news/claude-fable-5-returns-after-block/https://aitrustcert.com/en/news/claude-fable-5-returns-after-block/Claude Fable 5 is coming back globally after roughly two weeks of a block tied to export control and a safeguard bypass. We explain what happened, what Anthropic changed and why this story is a practical lesson about AI vendor risk for every company.Wed, 01 Jul 2026 00:00:00 GMThttps://aitrustcert.com/en/news/claude-sonnet-5/https://aitrustcert.com/en/news/claude-sonnet-5/Anthropic has introduced Claude Sonnet 5 - a faster, cheaper and more agentic model from its mid-tier Sonnet line. We explain why this launch is a signal about the direction of the whole AI market, and what organisational duties come with increasingly autonomous AI.Wed, 01 Jul 2026 00:00:00 GMThttps://aitrustcert.com/en/news/gpai-ask-your-vendor-about-compliance/https://aitrustcert.com/en/news/gpai-ask-your-vendor-about-compliance/The obligations of GPAI model providers aren't the end of the story. Even if you don't build your own model, you're responsible for how you use off-the-shelf AI tools. We explain what to ask your vendor: data, documentation, limitations, copyright and AI Act compliance.Thu, 25 Jun 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-vendor-risk-who-you-depend-on/https://aitrustcert.com/en/news/ai-vendor-risk-who-you-depend-on/The most advanced AI models have become strategic infrastructure - and access to them can be the subject of disputes and restrictions. We explain what AI vendor risk is, why a register of tools is the starting point and how to check who your company really depends on.Tue, 23 Jun 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-in-public-administration-sejm-nask/https://aitrustcert.com/en/news/ai-in-public-administration-sejm-nask/The Sejm and NASK are bringing order to AI in administration: rules, data, security and accountability. We show what companies can learn from it - and why AI under control starts with order in the organisation, not with technology.Fri, 19 Jun 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-literacy-first-line-of-defence/https://aitrustcert.com/en/news/ai-literacy-first-line-of-defence/AI literacy isn't a course on using ChatGPT, but the practical ability to use AI safely in everyday work. We explain why it's the first AI Act obligation, how to tailor it to roles and why training alone isn't enough without rules and evidence.Mon, 15 Jun 2026 00:00:00 GMThttps://aitrustcert.com/en/news/polish-ai-act-bill-in-parliament/https://aitrustcert.com/en/news/polish-ai-act-bill-in-parliament/Poland's draft Act on artificial intelligence systems has reached the Sejm - the AI Act stops being an abstraction and becomes a national oversight system. We explain what changes and what companies should prepare before inspections begin: from an AI inventory to evidence of action.Fri, 12 Jun 2026 00:00:00 GMThttps://aitrustcert.com/en/news/shadow-ai-biggest-gap-ai-act/https://aitrustcert.com/en/news/shadow-ai-biggest-gap-ai-act/AI is already running in many companies - just not always where leadership thinks. We explain what Shadow AI is, why it's the biggest gap in AI Act readiness, and where to start to regain control: from an inventory of tools to a register, usage rules and risk assessment.Fri, 05 Jun 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-act-not-only-for-tech-companies/https://aitrustcert.com/en/news/ai-act-not-only-for-tech-companies/“We only use AI, we don't build our own models” - a common but mistaken assumption. We explain who the AI Act really applies to: providers, deployers, importers and distributors - and why how and why a company uses AI matters too.Sun, 31 May 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-act-is-here-ai-governance-ai-literacy/https://aitrustcert.com/en/news/ai-act-is-here-ai-governance-ai-literacy/AI entered companies faster than the rules for using it responsibly. We explain why preparing for the EU AI Act should begin with two foundations - AI Governance and AI Literacy - and what practical steps a company can take first.Sat, 30 May 2026 00:00:00 GMThttps://aitrustcert.com/en/news/claude-opus-4-8/https://aitrustcert.com/en/news/claude-opus-4-8/Anthropic has released Claude Opus 4.8 - the most powerful model in the Claude family. But the most interesting thing about this launch is not 'faster and better'; it is the emphasis on honesty, admitting uncertainty and a lower tendency to let its own mistakes slip through. What does that mean for companies?Fri, 29 May 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-agents-what-they-are-why-companies-need-them/https://aitrustcert.com/en/news/ai-agents-what-they-are-why-companies-need-them/An AI agent is not just a model that answers a question - it is a system that understands a goal, plans steps, uses tools and carries out an action. We explain how it differs from a chatbot, where it makes sense, how to build one, and why it demands organisational maturity.Thu, 28 May 2026 00:00:00 GMThttps://aitrustcert.com/en/news/new-wave-of-llm-models-agents-and-trust/https://aitrustcert.com/en/news/new-wave-of-llm-models-agents-and-trust/The latest updates from OpenAI, Anthropic and Google point in one clear direction: LLMs are becoming agents that do not just respond, they act. What does this mean for organisations, and why is trust in AI more important than ever?Wed, 27 May 2026 00:00:00 GMThttps://aitrustcert.com/en/news/how-large-corporations-use-ai/https://aitrustcert.com/en/news/how-large-corporations-use-ai/AI in large corporations has moved beyond experimentation. It is now woven into the daily fabric of work. We look at how banks, retail, pharma and HR are using it - and why this matters for trust.Tue, 26 May 2026 00:00:00 GMThttps://aitrustcert.com/en/news/ai-act-august-2-2026/https://aitrustcert.com/en/news/ai-act-august-2-2026/Some dates travel far - through presentations, webinars and compliance checklists. 2 August 2026 is one of them. We explain what the date actually means for businesses, public bodies and organisations using artificial intelligence.Mon, 25 May 2026 00:00:00 GMT