Antivirus and anti-malware programs may protect your PC from known security threats, but it’s the unknown malicious software which gives attackers a chance to compromise your system.
According to a report by Engadget, new pieces of malicious software can infect a PC before its antivirus program has updated – or systems whose users do not regularly download the latest updates.
Security researchers are looking to combat this issue using deep learning algorithms that can identify new malicious code without database updates.
Israeli startup Deep Instinct plans to use an artificial neural network to build an antivirus suite that can recognise new malware with 20% higher accuracy than today’s protection software.
Microsoft and Invincea have also published papers on the potential of deep learning malware detection systems. One experiment found 95% of new malware without updates.
Deep Instinct’s antivirus solution is yet to be tested.