BasicsCover story
What is AI, really?
No formulas and no jargon. Here is what artificial intelligence is, how it seems to know so much, and why it sometimes gets things wrong.
Typing a message
Good
Technology, explained clearly and calmly.An information site for readers over 50.
BasicsCover story
No formulas and no jargon. Here is what artificial intelligence is, how it seems to know so much, and why it sometimes gets things wrong.
Typing a message
Good
Artificial intelligence, or AI, sounds like something out of the movies. In fact, you have probably been using it for years. When your phone suggests the next word as you type a message, that is AI. When your email moves advertising into a separate folder on its own, that is AI too.
In the last few years, programs have appeared that you can write to or talk with the way you would with a person. The best known are ChatGPT, Gemini, Copilot and Claude. People call them chatbots or AI assistants. You ask a question in everyday words and get an answer in everyday words.
Imagine someone who has read millions of books, articles and letters but has never left the library. They know what a good answer usually sounds like. But they have never been to the places they describe, and they can’t check whether something changed yesterday.
AI assistants work the same way. They were trained on an enormous amount of text and learned which words and ideas usually go together. When you ask a question, they build the answer word by word, each time picking the most likely next step. It is just like the suggestions on your phone, only far more skilled.
Not in the human sense. The program has no feelings or intentions. It can sound confident and even warm, but that is a way of writing it picked up from the texts it learned from. That is also why AI sometimes invents a fact, a date or a name and presents it with a straight face. Specialists call this a hallucination.
Treat AI as a well-read helper. For important decisions, check with a person you trust or an official source.
Practical
Five everyday situations where AI saves time and stress. For each one we give you a ready-made phrase you can type or say to the assistant.
A letter from the tax office, a phone contract, the leaflet inside a box of medicine. Copy the text or take a photo, then ask AI to explain what it says and what you are expected to do. You can ask again about any sentence that isn’t clear.
Careful: before you photograph a document, cover your ID number, address and account numbers.
“Explain in plain words what this letter says and whether I need to do anything by a certain date.”
A video call, photos for the grandchildren, a voice message. Each of these is easy once you have done it one time. An AI assistant walks you through it step by step and explains again if you ask a fifth time.
Tip: ask a relative to mark your most important contacts as favorites, so you can reach them with one tap.
“I want to video call my grandson on WhatsApp. Explain one step at a time and wait until I tell you I’m ready.”
Your phone can probably already read articles aloud, describe what is in a photo, and work as a magnifier for a restaurant menu or the small print on a package. These are AI features built into the phone itself.
Tip: most of these features live under Settings → Accessibility. Explore them once with someone close to you.
“Where is the magnifier on my phone, and how can I open it faster?”
The pill at 8 a.m., the dentist appointment, the grandchildren’s birthdays. The voice assistant on your phone can remind you of them, and a chatbot can write you a shopping list or a week of low-salt meals.
Careful: AI can organize your schedule, but never change a dose or a medicine without your doctor.
“Remind me every day at 8 p.m. to take my blood pressure medicine.”
A foreign language before a trip, the history of your hometown, gardening, the recipes you grew up with. AI answers at three in the morning and explains again as many times as you like.
Tip: if an answer seems odd, ask where the information comes from and check it somewhere else.
“Teach me 10 useful Italian phrases for travel and show me how to pronounce each one.”
Safety
AI makes scams more convincing. Here are four situations that are becoming more common, and what to do in each one.
The situationA call in a loved one’s voice: an accident, an arrest, money needed right now.
What to doHang up. Call the person yourself on the number you already have. Today a short clip from social media is enough to copy someone’s voice.
The situationA video of a famous person recommending an investment with “guaranteed returns.”
What to doIt is almost certainly a fake video, known as a deepfake. Guaranteed returns don’t exist. Don’t hand over your phone or install apps on a stranger’s instructions.
The situationA text or email that claims to be from your bank, a delivery company or the tax office, with a link to pay.
What to doDon’t tap the link. Call the number on the back of your card or the one on the organization’s official website.
The situationSomeone asks you to read out a code you just received by text message.
What to doNever share that code. Your bank and the police will never ask you for it.
Agree on a family password. Pick a word only you and your close family know. If someone calls claiming to be a relative in trouble, ask for it.
Think it’s a scam? Contact your bank right away and report what happened to the police.
Column
Readers ask, our adviser answers. No jargon, no rush.
Answered by our adviseron digital skills
My grandson tells me to ask ChatGPT. Is it free, and do I have to give my personal details?
AnswerThe basic version of most AI assistants is free. To sign up, they usually ask for an email address. Paid versions exist, but the free one is enough for everyday questions.
Don’t type in your ID number, card number, passwords or any health details you wouldn’t share with a stranger. If someone calls or writes to say you must pay a fee to access AI, it is a scam.
I got a voice message from my son saying he had been in a car accident and urgently needed money. It was his voice. Could it be fake?
AnswerYes. Today a short clip from social media is enough to imitate someone’s voice. Hang up and call your son yourself on the number you have. If he doesn’t pick up, call another relative.
Never send money under pressure, however urgent it sounds. If you suspect a scam, report it to the police.
Can AI tell me what to do about my high blood pressure?
AnswerIt can explain what the numbers on your monitor mean and what questions to ask your doctor. It can’t examine you and doesn’t know your medical history. Make decisions about medicine only with your doctor.
Here is a good way to use it: “Make me a list of questions for my appointment with the cardiologist.” That way you arrive prepared.
I’m afraid I’ll press the wrong thing and break my phone.
AnswerWith ordinary taps it is almost impossible to damage your phone. Usually an app simply opens, and you close it with the Back button.
The real risk comes from messages with links and calls asking for codes. Be careful there. Everything else, explore freely.
Glossary
Six terms you will come across in the news and in conversations with your grandchildren.