The brain and the heart of
hyper-personalised AI.
ALFI brings adaptive leadership, foresight and intelligence. EDIE brings empathy-driven insight and engagement. Together they form an equitable, accessible foundation for safe, hyper-personalised AI assistants — for families, educators, communities and organisations.
Executive partner for decision-making, planning, and optimisation. Structured, future-oriented, and precise — like a calm management consultant on call.
Emotional wellbeing guide, relationship nurturer and engagement enhancer. Warm, validating and intuitively human — speaks like a trusted mentor.
Built for safeguarding first.
Every assistant built on ALFI + EDIE inherits the same foundation: privacy, security, equitable access and adaptive governance — so the next assistant we build (like SAMU for GCSE study) starts safe by default.
Safeguarding & privacy
Role-aware access, auditable interactions, and consent-driven memory — engineered for children, families and regulated organisations.
Adaptive governance
ALFI orchestrates corporate, operational and PMO-style governance so assistants stay aligned with policy as they scale.
Equitable access
Designed for children, families, community groups, educators, and business and technology professionals — in the UK and internationally.
Meet SAMU
A hyper-personalised study companion for Samuel's GCSEs — built on ALFI + EDIE once the foundation is proven.
SAMU will guide Samuel across his Year 10 GCSE subjects with expert reasoning from ALFI and emotional support from EDIE — all behind the framework's safeguarding and privacy controls.
Coming after the foundational framework is live and validated.
AI Regulation & Governance Topics
Stay ahead of evolving global standards. Access clear, plain-English executive summaries on critical AI compliance frameworks compiled by Automated Governance Systems.
Eight Completely Prohibited AI Practices (Banned Since Feb 2025)
Entry into Force
The regulation officially became active law across the EU.
Prohibitions & AI Literacy
Bans on prohibited practices and mandatory AI literacy for staff became enforceable.
GPAI Model Rules
Rules governing General-Purpose AI models and member-state penalties take effect.
Standalone High-Risk AI
Compliance is required for credit rating, HR, and educational standalone AI systems.
Sectoral High-Risk AI
Regulated product-embedded systems (e.g., medical devices, vehicles) must comply.
No Exceptions for Location
Even if your office is in the UK, US, or Asia, if an EU citizen interacts with your AI or benefits from its output, you must comply fully.
Bans Are Active Now
Forbidden practices—like measuring student or employee emotions, profiling personalities, or social scoring—became illegal in early 2025.
SMEs & Startups Get Breaks
Fines for smaller companies are capped, and regulatory sandboxes are provided to help you test systems safely without fear of instant penalties.
No Hidden Chatbots
If a user is chatting with your system, you are legally required to make it blindingly obvious that they are speaking with an AI, not a real human.
List all AI systems our business develops, deploys, or purchases from third parties.
Confirm none of our active tools perform social scoring, employee emotion tracking, or subliminal manipulation.
Determine if any AI is used in hiring/recruitment, student evaluation, credit underwriting, or critical safety systems.
Ensure every chatbot, text writer, or media generator clearly states to end-users that they are interacting with AI.
Ensure third-party AI suppliers provide full technical specs, accuracy guidelines, and cybersecurity guarantees.