Artificial intelligence is shaking up pretty much every corner of modern business, and cybersecurity is no exception. Of course, it’s sadly not just the good guys using it. Hackers are now using AI to launch faster, smarter, and scarily believable attacks, while cybersecurity teams are fighting back with AI-powered defences of their own.
How Hackers are using AI
Phishing emails used to be easy to spot. Poor grammar, strange phrasing, and suspicious links were obvious giveaways. It’s likely that we were all contacted by a “Nigerian Prince” offering to give up his riches at some point! Nowadays, these types of scams are basically prehistoric. AI tools can now generate emails and more that look entirely legitimate, grammatically correct, professionally written, and even tailored to specific regions or industries.
Emails
An email that appears to come from a known supplier, complete with correct branding and familiar language, may now be the product of an AI model trained to imitate that company’s tone and writing style. The result is a message that looks right, sounds right, and can easily bypass a distracted employee’s attention.
Deepfake voices
AI-generated deepfake voices are making this even more dangerous. With just a few seconds of recorded speech, say from a podcast, video, or even a voicemail, attackers can clone a person’s voice and use it to impersonate senior staff. A single fake call could be enough to authorise a payment or share sensitive information.
Fake websites
Meanwhile, entire fake websites can now be created automatically using AI-based web generators. These cloned sites often look identical to legitimate ones, complete with logos, contact details, and even working login pages. Once a user enters their credentials, they’re handed straight to the attacker.
This level of automation and sophistication means cyberattacks are no longer limited by time or skill. For many smaller businesses without dedicated security teams, distinguishing real from fake has become significantly more difficult.
How defenders are responding
It’s not all doom and gloom though! Fortunately, the same technology that enables these attacks is also strengthening the defences. AI-powered cybersecurity systems are now capable of detecting and blocking threats long before they reach users.
Modern email filters use AI to “sandbox” suspicious links and attachments, safely testing them in isolation before they’re delivered. The process takes milliseconds and happens behind the scenes, drastically reducing the chances of a single click causing widespread damage.
Advanced systems also analyse communication patterns to detect subtle irregularities. They combine traditional authentication checks, such as SPF, DKIM, and DMARC, with AI-driven behaviour analysis to identify when something doesn’t match normal activity. You can find out what the heck all those letters mean and how they relate to email domains by giving our previous blog a read!
AI doesn’t just protect your inbox either. It can also stand guard over your network too, spotting anything unusual as it happens so you can fix it before it causes real damage.
AI vs AI: What is the local impact?
Certain industries, like finance, manufacturing, and legal services, are feeling the pressure more than most. They deal with sensitive data and financial transactions every day, which makes them obvious targets for AI-powered attacks.
Local cybersecurity specialists are warning that traditional antivirus tools and spam filters are no longer enough. AI has dramatically increased the scale and speed of cybercrime. For example, a single attacker can now generate thousands of personalised phishing attempts or fraudulent websites in mere minutes.
What businesses in Birmingham and beyond can do right now
Sadly, there’s no single solution to stop AI-driven attacks, but there are a few things you can do right away to protect your business:
Upgrade email security
Use solutions that scan links and attachments, and authenticate sender domains (SPF, DKIM, DMARC).
Train staff regularly
Employees remain the first and most important line of defence. Training should cover phishing awareness, deepfakes, and verification practices.
Verify unusual requests
Implement a “trust but verify” policy for all financial or data-related actions.
Adopt AI defensively
Use AI-based monitoring tools to detect anomalies quickly and automatically.
Stay informed
Keep up with guidance from regional cybersecurity networks and resilience centres – you’ve just done this one by reading this blog, well done!
FAQ Section
What does AI in cybercrime actually mean?
It refers to hackers using machine learning to craft convincing phishing messages, voice calls, or websites that mimic real businesses.
How does this affect a Birmingham SME?
If you run a law firm in Colmore Row, for instance, your staff could receive an AI-generated email that looks identical to one from a regulator.
What does a real-world AI attack look like?
You might get a call that sounds exactly like your MD asking for a payment but it’s a cloned AI voice.
How can businesses fight back?
By deploying AI-driven filters, enforcing MFA, and training staff to verify unexpected requests.
Summary (TL; DR)
- Hackers use AI to make phishing and scams more convincing.
- Defenders use AI to scan, sandbox, and block suspicious emails.
- Birmingham SMEs are at rising risk as AI-driven crime scales up.
- Awareness and AI-based tools are the only way to keep pace.
- Staff training on deepfakes and email hygiene is essential.
Need some extra help with keeping your business cybersecure? Book a free, no obligation chat with Mark to talk about how we could work together!