Outspoken artificial intelligence expert Hassan Taher sees the good, the bad, and the ugly of the ever-evolving AI technology landscape.
"I believe that AI has the power to make a significant impact on people's lives," mentioned Taher in a recent interview while discussing the impact of AI on health care. That's the good part.
Hassan Taher also recognizes the dangers that AI technology can germinate, especially in matters of cybersecurity. For this reason, he urges organizations that grow their AI footprint to tread carefully. AI has led to rapid innovations and expansion across many industries, Hassan Taher explains. But as those expansions and innovations cause additional growth, the possibility of cyberattacks grows at the same time. For a range of specific cyberattack types, including phishing, malware, and spoofing, the ability of AI to be manipulated to facilitate cyber attackers' actions while multiplying their efforts cannot be ignored.
Hassan Taher Weighs AI's Risks and Rewards
That risk cannot be allowed to outweigh the benefits that AI brings to society, however. Hassan Taher recognizes that in many cases, AI is able to pick up on patterns in data that are often missed by human analysis simply due to the sheer overwhelming volume of data that is available, like noticing issues that may turn into recalls with consumer products, specific imaging results indicating disease, or similar data points that are often lost in the noise of overwhelming volume.
For this reason, a balance must be struck between usage and security. AI has the capability to cause serious ethical issues in cyberattacks by blurring accountability borders while raising questions about the ethical issues that may arise from using AI. To remedy this, Taher suggests AI development needs to follow specific ethical guidelines or regulations that help prevent misuse. These include several cybersecurity threats that must be addressed as the technology continues to roll forward.
Phishing is a type of attack that uses fake emails or websites to trick individuals into providing hackers with sensitive details, Hassan Taher notes. Because AI is capable of generating very convincing fake websites and emails, it becomes easier to trick people into falling for a phishing scheme. Similarly, AI is also able to analyze vast amounts of information to customize its phishing attempts to specific targets it has identified, making the attempts more convincing. As one example, DeepPhish is an AI-powered approach that emerged in 2018, wherein cybercriminals leveraged machine learning to learn legitimate website patterns and then recreate them so the phishing websites they created were very believable.
Another approach that Hassan Taher cautions about is malware. As technology advances, malware, or malicious software that is designed to exploit security holes in a device, is becoming more sophisticated. Because AI is capable of learning and adapting, it can also create malware that is able to mirror this capability. AI-designed malware can then evade countermeasures and detection in a more effective manner. As an example, MyloBot, which was discovered in 2018, was a botnet that used sophisticated algorithms developed by AI to take over computers, avoid detection, and use compromised computers to launch other cyberattacks or further distribute the malware.
The final cybersecurity concern that Hassan Taher is concerned about is spoofing. An approach that impersonates individuals and tricks other victims into taking action, spoofing could be advanced through the use of AI to create very convincing fake video and audio content known as deepfakes. A CEO at a UK-based energy firm was mimicked in 2019 with an AI-generated voice, even including the CEO's unique voice patterns and German accent. This led the firm's financial controller to think that the request for a funds transfer totaling $243,000 was genuine, causing significant harm both to the company's financials as well as its public reputation when the spoof was revealed.
Taher Says AI Adapters Need More Than an Ounce of Prevention
But what approach should be taken to prevent these issues from becoming problematic? With AI technology's ability to evolve, as well as innovations in the cyberthreat landscape, Hassan Taher knows that staying updated and informed about new developments is the most important step to preventing these issues. Three ways to prevent these issues is regularly updating both software and hardware, staying educated about the risks of cyberattacks, and using strong, unique passwords.
Hassan Taher notes that the process can't rely entirely on individual scrutiny and action, however. Businesses and governments can counter the threat posed by AI-powered cyberattacks by investing in a range of AI-powered cybersecurity solutions. This allows these organizations to prevent, identify, and respond to cyberattacks as they happen by leveraging the power of AI against malicious parties. By incorporating AI into the process, the same data can be noted to help predict and prevent future threats, while at the same time developing effective countermeasures against these attacks.
Because cyberattacks cross borders, it's vital that global ethical guidelines for AI be developed and international cooperation increased to help prevent AI from being misused. Hassan Taher notes that countries could work together to develop international regulations prohibiting AI usage for malicious intentions and take action to enforce regulations across borders on the international stage.
Though AI does have the capability to advance and, in turn, increase the risk posed by cyberattacks, undertaking appropriate measures such as AI-powered cybersecurity, remaining informed about potential risks, and improving international cooperation combined with ethical international AI guidelines and regulations are able to mitigate these possible risks.
As a brilliant expert and author on the possibilities of AI in humanity's future, Hassan Taher is at the forefront of digital transformation using AI to increase our capability. He recognizes that though there are threats to this bright and promising future, AI is here to stay and provides great opportunities to improve the lives of everyone on our planet, bringing us to a brighter future together.
