Last updated: 21 October 2026
Like our AI overview from July, this is an article with a short shelf life. Fraud methods change from month to month, and the figures below are current as of 21 October 2026. Read it as a snapshot, not an answer key. The point is not to frighten, but to give an honest picture of what is actually happening right now, and what you can do about it.
October is National Security Month in Norway (Nasjonal sikkerhetsmåned), and this year's theme is "Digital judgement – small choices, big consequences". It is a fitting theme. The headlines tend to be about sophisticated attacks, but for most of us it comes down to a small moment: whether we tap the link, hand over the code, or hang up and call back ourselves.
The figures, and some good news
According to a fresh survey from NorSIS, the Norwegian Centre for Information Security, close to 500,000 Norwegians have been defrauded of money online in the last two years. Around 100,000 have had their BankID, the Norwegian electronic ID, misused, and close to 300,000 have paid for goods or services they never received. Seven in ten do not believe, or are unsure, that they can recognise text and audio made with artificial intelligence.
That is the bad news. The good news is that the defences are working. Since November 2024 the Norwegian telecom operators have stopped fraudsters from borrowing Norwegian phone numbers, and according to the Norwegian Communications Authority (Nkom) more than 100 million fraud attempts using Norwegian numbers have been stopped over the last two years. According to figures from the Financial Supervisory Authority, fraud losses at the banks fell by NOK 265 million from 2024 to 2025. Nkom is honest that nobody can say exactly how much each measure contributed, but the direction is right.
The fraudsters have responded by adapting. Telenor sees more misuse of foreign numbers, and fraudsters trying to move the conversation over to WhatsApp or Telegram, where the operators' filters do not reach. Artificial intelligence has not created new fraud, Telenor says. It has made the old methods far more effective.
How fraudsters are trying to fool you right now
The methods are few, but they are carried out better and better. These are the ones that keep coming up this autumn.
"We will help you stop the fraud." Someone calls and says they are from PayPal, the bank or a public agency. Payments have been made in your name, and you need to help stop them. To do that you have to install an app, for example AnyDesk or TeamViewer, and at that moment a stranger has access to your phone. The insurer Frende warned in August that fraudsters now take the money via Apple Pay from your card, so you never get a BankID request. Many people have learned that BankID is the warning light. The fraudsters know that, and have set the scam up so that the light never comes on.
A knock at the door. Finance Norway warns of fraudsters who pose as police officers, bank staff or technicians, and who may turn up at your home. It sounds too brazen to be true, but it happens. Your bank does not send anyone to your home to collect your card, code device or BankID.
A text message from an agency you know. The Norwegian Public Roads Administration (Statens vegvesen) registered 770 fraud attempts in its name from June to September, 48 per cent more than the year before. Fraud by text message rose from 48 to 206 cases in one year, and September was the worst month so far. Almost half concern the driving licence, and the AutoPASS toll system comes second. The agency's advice is simple: go straight to vegvesen.no instead of tapping the link.
A friend who needs a code. You get a message from a friend on Facebook: the two of you can win something together, but you need to enter a code you receive by text message. The code is in practice a temporary password to your account, and once you send it, the account is no longer yours. Afterwards the same message goes on to your friends, from your account. On Instagram the variant is to ask for a screenshot of a text message containing a link.
"Hi mum, I have a new phone." A message from an unknown number: your child has lost their phone and needs money, and it is urgent. DNB writes that AI is now often used to manipulate the voice in this kind of fraud, while Telenor says that advanced deepfakes are not yet widely used in the fraud it sees. The advice is the same either way: call your child's usual number, and ask a question only your child can answer. Do not write your child's name yourself. It only helps the fraudster.
What this means for a business
Businesses have more to lose, and the fraudsters know it. DNB, Norway's largest bank, reports that the number of business customers who were defrauded rose by 35 per cent in the first half of the year. In July alone, the bank detected ten targeted attacks on people with access to company accounts, with a combined attempted amount of around NOK 10 million. What sets them apart from ordinary fraud is that the attacks are tailored. Where ordinary phishing casts a net, these attackers use a hook.
Artificial intelligence makes this harder to spot. The insurer Tryg warned 43,000 Norwegian business customers about CEO fraud this summer, pointing out that fraudsters now write flawless Norwegian, copy language and tone, and in some cases imitate both voice and appearance. How far it can go was shown by an Italian private bank this autumn. According to Reuters, the chair of Fideuram received a WhatsApp message that appeared to come from the group CEO, followed by a phone call with an AI-cloned voice. The result was transfers of 95 million euros, of which only around 53 million has been traced.
What stops this is not technology first and foremost. It is routines:
Two people approve every payment. One person can be fooled. Two people who both have to say yes are far harder to fool at the same time.
Call back on a number you already have. If a supplier asks to change their account number, or "the boss" asks for an urgent transfer, call back on a number you already know. Never on the number in the email.
Urgency is a warning, not a reason. Tryg puts it well: the more urgent it is, the more important it is to stop and check.
Make it safe to speak up. In the NorSIS survey of employees at small businesses, close to three in ten are unsure whether they can handle digital threats, and 19 per cent are unsure whether they can report an incident without being blamed. An employee who is afraid to say "I think I clicked on something wrong" waits too long to say it.
When the services we rely on fail
Not everything is about fraud aimed at you personally. The Norwegian Data Protection Authority (Datatilsynet) received 282 reports of personal data going astray after deliberate attacks in the first half of this year, against 103 in the same period last year. The most important explanation, according to the authority, is that a single attack can hit many organisations at once, because they use the same service.
On Wednesday 21 October we got an example of how fast this can move. Two critical holes in PHP, the software that runs a large share of the world's websites, led the Norwegian Digitalisation Agency, the education login service Feide and the University of Stavanger, among others, to take websites down while they updated. The Norwegian National Security Authority (NSM) urged all system owners to update immediately, but at the time of writing was not aware that the holes had been exploited.
For an ordinary user this means little more than some downtime. For a business with its own website the question is simpler than it sounds: do you know who updates it, and how quickly? We have written more about why hosting and maintenance got more important, and this Wednesday is a good illustration.
Five habits that stop most of it
Hang up, and call back yourself. Whether it is the bank, the police or a public agency: call back on a number you have found yourself, never on a number you were given in the call or the message.
Never share codes. Not BankID, not text-message codes, not with anyone. A code someone asks you for is a key to something that is yours.
No apps and no screen sharing on request. No serious bank or agency asks you to install a program to "stop fraud".
Turn on two-step login. NorSIS found that 41 per cent do not use it where it is available. It is one of the simplest improvements there is.
Agree on a code word in the family. A word only you know exposes a fake "hi mum" or a cloned voice in seconds.
Where are we, really?
Briefly summarised in October 2026: the defences are working, and losses are going down. But the fraudsters have better tools, and they use them to make the same old tricks more convincing. What has changed is not what they try, but how real it looks.
That is why this year's theme for National Security Month fits so well. It is rarely the technology that fails when someone is fooled. It is a small choice, made in a moment of urgency: to tap, to hand over, to install. The most important habit is also the simplest. Stop, and call back yourself.
Wondering how well your business would stand up to this? Feel free to have a no-obligation talk.