Cybersecurity experts have urged caution with over-reliance on AI in cyber defence in a new Finextra Research paper exploring predictions for cybersecurity in the year ahead.
In The Future of Cybersecurity: 2020 Predictions, published by Finextra Research, partner at EY Steve Holt points out the danger of cyber criminals targeting machine learning processes and embedding fraudulent mechanics into AI engines.
"One of the big concerns is the underlying integrity of the data," Holt says. "So, if the attackers don’t do big enormous payouts immediately but attempt to alter the underlying data, how would that be spotted?”
For this reason, caution is urged with financial services becoming overly bullish with the use of AI to address their cyber-related concerns.
2020 may therefore be another year of increased use of AI in cybersecurity, but big developments may only come further down the line, according to Dries Watteyne, head of SWIFT's cyber fusion centre.
The report goes on to highlight risks that emerge as the financial world becomes more interconnected, as banks partner with companies with less sophisticated protection than their own or move their data to cloud services provided by tech giants.
"Looking ahead to 2020, attackers will continue to really test cloud providers as they know how much more data they are hosting," Holt says.
The report also explores the evolution in motives of cyber attackers, going from pure financial gain to disruption and protest, and predicts expanded use of ethical hacking by financial organisations to stay ahead of the curve with how cyber criminals are likely to operate.
Click here to download 'The Future of Cybersecurity: 2020 Predictions'.