Pyralink™ Prove It / Secure AI Toolkit
Free toolkit
The Secure AI Toolkit
Free, printable resources for anyone who has to make AI and cloud use safe: compliance and risk, IT, security leads and founders. Written in plain English, with a link to the primary source for every claim. No sign-up needed.
Nine resources
Pick the one that matches the job in front of you. Every page prints cleanly (use your browser's print option to save as PDF) and has a Markdown download.
Set the rules
AI acceptable-use policy: a starter template
A plain-English, editable starter template for an organisation's AI acceptable-use policy, with notes on what to change. Free to copy. Not legal advice.
Secure AI adoption checklist
A printable checklist for adopting AI safely: shadow AI discovery, data classification, prompt and data-leak controls, vendor due diligence, access control, logging, human oversight and incident response.
AI vendor security questionnaire
A free questionnaire for assessing the security and data handling of an AI supplier, with what a good answer looks like and the red flags to watch for. Printable and downloadable.
Defend and respond
Prompt injection and data leaks: a defence guide for staff
A plain-English guide for staff and IT on prompt injection and data leaks when using AI tools and agents: what they are, what to never paste, what to check, and what builders should do. Printable.
Cloud security basics checklist
A free, printable checklist of the cloud security basics that matter most, for IT teams and small organisations using AWS, Azure, Google Cloud or SaaS. Based on NCSC guidance.
AI incident response mini-playbook
A short, printable playbook for responding to AI-related security incidents, such as data pasted into a public tool, a prompt-injected agent, a compromised AI package or a deepfake payment request. For small and mid-size teams.
Stay informed and protected
UK and EU AI regulation and guidance tracker
A dated tracker of UK and EU AI regulation, standards and official guidance that matters to security, IT and compliance teams. Every date links to its primary source. Last reviewed 31 October 2026.
AI and cloud security glossary
Plain-English definitions of AI security, cloud security and AI regulation terms used by compliance, IT and security teams, with links to primary sources.
Protect yourself from AI scams: deepfakes, voice clones and AI phishing
A practical guide for individuals and families on AI voice-clone and deepfake scams, AI-written phishing, and simple verification habits that work. Printable. Based on UK official advice.
Primary sources we rely on
The toolkit is written in our own words and points you to the original documents. Read them: they are the authority, and we are not.
- NCSC: National Cyber Security Centre guidance on secure AI development, cloud security, phishing and incident management
- UK DSIT: AI Cyber Security Code of Practice
- ICO: Information Commissioner's Office: personal data breaches and AI guidance
- EU AI Act: Regulation (EU) 2024/1689, official text on EUR-Lex
- EU AI Act timeline: European Commission AI Act Service Desk implementation timeline
- NIST AI RMF: NIST AI Risk Management Framework
- ISO/IEC 42001: ISO/IEC 42001:2023, AI management systems
- OWASP: Top 10 for LLM Applications
- MITRE ATLAS: Adversary tactics and techniques for AI systems
- CISA and partners: Careful Adoption of Agentic AI Services (1 May 2026)
Using the toolkit
Not legal advice. These resources are general information and starting points. They do not take account of your sector, contracts or risk appetite, so adapt them and have the important parts reviewed by someone qualified.
You may copy, adapt and print them for your own organisation. Please do not republish them as your own, and keep the credit to Pyralink™ Prove It. Spotted a mistake or out-of-date item? Email mike.ade@pyralink.co.uk.
Pyralink also makes PyraLink™ Prove and AI Paste Guard. The toolkit is written to be useful whether or not you ever use them.