AI assistants such as ChatGPT, Google Gemini, Microsoft Copilot, Meta AI (inside WhatsApp) and Claude can write, explain, summarise and translate in seconds. Used well they save hours. Used carelessly they produce confident mistakes and can expose private information.
Good uses
- Drafting: a first version of a letter, a CV summary, a business proposal, a speech, a social media post. Then you edit it into your own voice.
- Explaining: "Explain compound interest like I am 15", "What does this clause in my tenancy agreement mean?"
- Summarising a long document or email thread.
- Planning: a study timetable, a budget, a menu, an event checklist.
- Practising: mock interview questions, language practice, quizzes before an exam.
- Fixing: "Rewrite this politely", "Make this shorter", "Check my grammar".
How to ask
The more context you give, the better the answer. Compare:
- Weak: "Write a cover letter."
- Strong: "Write a one-page cover letter for a junior accountant role at a bank in Accra. I have an HND in Accounting and did national service at the Ghana Revenue Authority. Keep it formal but warm. Here is the job advert: …"
Then refine: "Shorter", "Less formal", "Add my experience with QuickBooks", "Give me three options".
Check what it tells you
AI assistants can state wrong things with complete confidence, including invented facts, quotes, statistics and references.
- Check any fact that matters against an official source: fees, deadlines, laws, medical and financial information.
- Local details, such as Ghanaian government procedures, prices and phone numbers, are where they most often go wrong or out of date.
- Ask it to show sources, then open them. If a link does not exist, neither does the fact.
Never paste these in
- Passwords, PINs, and one-time codes.
- Your full Ghana Card number, bank or MoMo account details.
- Other people's private information, medical records, or confidential documents from work, unless your employer has approved the tool for that.
Assume anything you type may be stored. Check each service's privacy settings; several let you turn off the use of your chats for training.
At school and work
Many schools and employers have rules about AI. Using it to understand and to draft is widely accepted; submitting its work as your own often is not, and teachers increasingly recognise it. When in doubt, ask, and say when you have used it.
Scams using AI
AI makes fake voices, videos and messages more convincing. A voice note that sounds like a relative asking for money, or a video of a public figure promoting an investment, can be fake. The old rule still works: verify by calling back on a number you already know. See spotting a fake message.