GET AI MENA
Insights
Notes from our work on AI governance, adoption and delivery across Jordan and the GCC.
All articles
AI Adoption Metrics: A Checklist That Avoids Vanity Numbers
Which usage and outcome measures show that AI licences changed real work, which vendor numbers mislead, and a one-page dashboard with a decision rule per tile.
Read the article ↗
RAG vs fine-tuning: when retrieval, training or a better prompt wins
Retrieval changes what a model can see; fine-tuning changes how it behaves. A comparison of cost, updates, citations and upkeep, with a decision table and a prompt test to run first.
Read the article ↗
Is ChatGPT safe for business data? Plans, settings and paste rules
Personal and business ChatGPT plans handle your data differently. What changes on training, retention, admin access and location, plus a one-page rule for what staff may paste.
Read the article ↗
AI regulations in the Middle East: what binds you in 2026
Which AI rules bind in Saudi Arabia, the UAE, Jordan, Qatar and Egypt as of October 2026, which only guide, and a monthly routine to track what changes next.
Read the article ↗
Fractional CTO: When You Need One and What the First 90 Days Hold
When does a business need a fractional CTO instead of a full-time hire? Our answer, a 90-day plan of what they own, and the measures that show whether it works.
Read the article ↗
Arabic OCR for scanned forms and invoices: what breaks and how to test it
Arabic OCR can work on scanned forms and invoices, but only after you test it on your own documents, field by field. What breaks, how to score accuracy, and where a person should check.
Read the article ↗
AI Risk Register: Columns, Owners and a Sample Row to Copy
A working AI risk register has one row per AI use, a named owner per risk and a review date people keep. Here are 14 columns, three roles, a review rhythm and a sample row.
Read the article ↗
n8n vs Zapier vs Make: hosting, data location, pricing and AI steps
A neutral comparison of n8n, Zapier and Make on hosting, where data is stored, how each one bills, and their AI agent features, with a table of who each tool fits.
Read the article ↗
Prompt writing for business users: context, examples and checks
How staff write AI prompts for real work: the context to give, examples to show, checks to run, a template to copy and a shared role prompt library.
Read the article ↗
AI voice agent for call centres: dialects, delay, handoff and recording
How to launch an Arabic AI voice agent on the phone: which calls to start with, testing dialects, cutting delay, handing over to people and the recording rules in Jordan, Saudi Arabia and the UAE.
Read the article ↗
Claude Enterprise rollout: SSO, data settings and first workflows
A step-by-step Claude Enterprise rollout: identity and SCIM first, retention and training settings next, shared projects before connectors, then two or three reviewed workflows.
Read the article ↗
Saudi PDPL and AI Tools: Lawful Basis, Transfers and the DPO
What the Saudi PDPL means when staff paste data into AI tools and when AI vendors serve Saudi clients: lawful basis, transfers abroad, the DPO and the 2026 draft changes.
Read the article ↗
AI Innovation Workshop: A Three-Day Sprint That Ends in Ranked Pilots
How to run an AI innovation workshop that ends in ranked pilot candidates: who to invite, a three-day agenda, a scoring table and a one-page pilot card.
Read the article ↗
Legacy system modernisation with AI: retain, wrap or rebuild
Old spreadsheets and databases do not always need a rewrite in one go. A decision table for when to keep, wrap or rebuild them, and where AI helps first.
Read the article ↗
AI agent skills: how to turn a written procedure into one
Staff procedures rely on judgement an agent does not have. How to rewrite one as a skill, with a worked supplier-invoice example and test questions that keep it working.
Read the article ↗
AI Impact Assessment (ISO/IEC 42005): A Template That Changes Decisions
An AI impact assessment should end in go, change or stop. Here is a copyable nine-part template, five screening questions, and a rule for who should sign.
Read the article ↗
AI Champions Program: How to Pick Champions and Keep Them Going
Champions keep AI in daily work after training ends, if you pick them for coverage and credibility and give them time. A scorecard, an 8-week plan and what to measure.
Read the article ↗
Arabic RAG: how to build a knowledge assistant that cites its sources
A step-by-step guide to Arabic RAG: OCR checks, spelling normalisation, chunking, hybrid search, cited answers, refusals and dialect tests, plus when not to build one.
Read the article ↗
LLM monitoring in production: a checklist for operations teams
Six signals to watch once an AI assistant or agent is live, who owns each alert, how to move to a new model before the old one is retired, and what to do in the first hour of an incident.
Read the article ↗
Measuring AI ROI: A Worked Example and Worksheet
Hours saved are not value until they are redeployed. A copyable worksheet with illustrative numbers in Jordanian dinars, from baseline to payback.
Read the article ↗
AI governance training for risk, legal and audit teams: a curriculum
A practitioner curriculum for risk, legal, compliance and audit teams: six modules, case exercises set in KSA, UAE and Jordan, and an honest way to assess.
Read the article ↗
Generative AI security risks: the controls and who owns them
What changes when assistants and agents read your email and files: six generative AI security risks, the control that reduces each one, and who should own it.
Read the article ↗
AI workflow automation: which workflows to automate first
A decision guide with a scoring table: which workflows to automate first, where human checkpoints belong, and how the automation should connect to your CRM and ERP.
Read the article ↗
First 100 days after acquisition: secure, stabilise, then add AI
A 100-day plan for the new owner of a software or AI-using business: who secures access, data rights and model contracts first, what to stabilise, and when to improve with AI.
Read the article ↗
Arabic AI Chatbot for Customer Service: A Practical Guide
How to build an Arabic AI chatbot for customer service on WhatsApp and the web: handle dialects, hand over to people, and measure resolution, not just containment.
Read the article ↗
AI Centre of Excellence: Central, Federated or Hub-and-Spoke?
Three ways to run an AI centre of excellence compared on decision rights, speed, cost, skills and risk, with the model we recommend for a ministry, a bank and a family group.
Read the article ↗
AI Proof of Concept: A One-Page Charter With Kill Criteria
A one-page AI proof of concept charter to copy: hypothesis, success and kill criteria locked before work starts, data allowed, time box, decision owner and go/no-go date.
Read the article ↗
AI Briefing for Boards: What Directors Need to Decide
An AI briefing for boards should end in decisions. Six decisions for directors, a table of evidence to request and a 90-minute agenda you can use at your next meeting.
Read the article ↗
Fractional Chief AI Officer: When It Fits and How to Measure It
Should you hire a full-time chief AI officer, form a committee or bring in a fractional one? Our decision guide, with a six-month scorecard.
Read the article ↗
AI coding assistants for engineering teams: a rollout with review gates
A five-step rollout for Claude Code, Cursor, Copilot or Codex: shared conventions, permissions in configuration, review gates, secrets handling and delivery metrics.
Read the article ↗
Microsoft Copilot Rollout Plan: Readiness, Permissions, Pilots
How to plan a Microsoft Copilot rollout: check readiness, clean up permissions before licences go out, run a pilot with named roles, and measure real use.
Read the article ↗
AI Readiness Assessment: How to Score It Honestly and Plan a Roadmap
Score data, people, process, governance and technology from evidence, rank use cases by value and readiness, and know when to skip the assessment and start with one workflow.
Read the article ↗
Vibe coding security: checklist for Lovable, Bolt or Cursor apps
What to check before an AI-built app takes real customer or staff data: access rules, exposed keys, Supabase RLS, prompt injection, backups, Arabic, and when to rebuild.
Read the article ↗
AI Acceptable Use Policy: What to Include, With a One-Page Template
A usable AI acceptable use policy fits on one page: approved tools, data rules, human review and a route to ask. Here is what to include and a template to copy.
Read the article ↗
AI adoption strategy: why rollouts stall and a 30-day recovery plan
Licences bought, usage flat after a month. Why enterprise AI rollouts stall, and a 30-day plan to pick workflows, appoint champions, measure real use and retire idle seats.
Read the article ↗
AI agent governance: guardrails for agents in real workflows
How to put AI agents into real workflows with enforceable limits: autonomy levels, tool permissions, human confirmation, audit logs, budgets, a kill switch and Arabic testing.
Read the article ↗
AI due diligence: what to check before you invest in an AI company
What to check before you invest in or acquire an AI company: model and data rights, third-party API dependency, evaluation evidence, security, data protection, team and cost to serve.
Read the article ↗
AI training for employees: role-based learning built on daily work
The effect of generic prompt webinars often fades within weeks. Role-based AI training starts from each role's real tasks, practises on redacted work in Arabic and English, and measures what changes.
Read the article ↗
Arabic LLM evaluation: how to test a model before you roll it out
A practical method to test whether a model or AI assistant works in Arabic: your own tasks, dialects and Arabizi, normalised scoring, a calibrated judge and a release gate.
Read the article ↗
Choosing an enterprise AI platform: Claude, ChatGPT, Copilot or Gemini
A vendor-neutral way to compare Claude, ChatGPT Enterprise, Microsoft Copilot and Gemini: where your documents live, residency, admin controls, Arabic, agents and cost structure.
Read the article ↗
Sovereign AI: deciding where your AI runs in Saudi Arabia and the UAE
Sovereign AI is a decision about where models run and where data travels. We compare managed APIs, in-country cloud and on-premises open-weight models under KSA and UAE rules.
Read the article ↗
A Practical AI Governance Framework for Saudi Arabia, the UAE and Jordan
An AI governance framework sets who decides on AI, which rules apply and how risk is tracked. This is a practical model for organisations in Saudi Arabia, the UAE, Jordan and the wider GCC.
Read the article ↗
ISO 42001 Certification: What the Standard Requires and How to Prepare
ISO 42001 certification confirms that an AI management system meets ISO/IEC 42001:2023. What the standard asks for, a readiness checklist, and what auditors check at each stage.
Read the article ↗
Shadow AI: How to Find and Govern the AI Tools Your Staff Already Use
Shadow AI is the AI your staff use without approval. Find it with surveys, SSO, CASB and expense data, then govern it with an approved-tools list and access and DLP controls.
Read the article ↗