Open Navigation

IEEE.org IEEE Xplore Digital Library IEEE Standards IEEE Spectrum More Sites

IEEE Transmitter Facebook Twitter Instagram LinkedIn

Open Search
Open Navigation
Open Search

Safeguards or Threats to Cybersecurity?

AI and quantum are a dynamic duo: AI enhances decision-making and efficiency and can also be used to help protect systems. Quantum computing solves complex problems. Yet, these technologies can also introduce new vulnerabilities in our cyber defenses.

Quantum Computing

Quantum computers leverage quantum mechanics to quickly tackle problems traditional computers can’t. Despite their potential, researchers are still working to overcome major challenges. When they do, quantum technology could revolutionize industries.

Check out the impact of Quantum

AI and Cybersecurity

AI revolutionizes sectors by making autonomous decisions, predicting outcomes, and optimizing resources. Ensuring AI system integrity and security is crucial as threats evolve.

Check out the impact of AI and Cybersecurity

Quantum Computing
AI and Cybersecurity

Quantum computers use quantum mechanics to process vast amounts of data along multiple computational paths. They aren’t ready for prime time. Qubits, the building blocks of quantum computers, are sensitive to outside interference, making it hard to get accurate results. Integrating many qubits into a single system is also difficult.

“With new materials, circuits, and algorithms being announced regularly, quantum computing enables a variety of next-generation services and applications, from sensing to communication to computing to simulation. A near-term breakthrough is highly likely.”

Houbing Song, IEEE Fellow

Quantum Breakthrough

Because they can process multiple computational pathways at once, quantum computers are particularly good at calculating solutions to problems that would take today’s classical computers years to solve. Quantum computers are expected to excel at problems like decrypting data, scheduling flights, drug discovery and pinpointing optimal shipping routes.

“Quantum computing is leaving laboratories and being applied to real problems in large companies, mainly to discover new materials and medicines. In the short term, the chemical and pharmaceutical industries should benefit most from quantum computing.”

Euclides Chuma, IEEE Senior Member

Preparing for Y2Q

When you send an email or shop online, it is encrypted, protecting your privacy and enabling the transaction. But there's a race to develop post-quantum encryption methods before “Y2Q,” the point in the next five years when quantum computers may be able to decrypt data previously thought to be secure.

“The threat of sophisticated actors storing encrypted data now to decrypt it later using quantum computers — a strategy known as "harvest now, decrypt later"— is a realistic possibility. To counteract this threat, businesses can take proactive steps such as adopting post-quantum cryptography well ahead of the full development of quantum computing.”

Kevin Curran, IEEE Senior Member

Quantum Showdown

Cybersecurity is often seen as an escalating war between attackers and defenders. Will quantum computers be yet another tool that benefits both sides?

“In cryptography, attack and defense is always a cat and dog fight. We improve a defense technique; the attacker will come up with a new method.”

Rahul Vishwakarma, IEEE Senior Member

AI is driving advancements across various sectors by enabling machines to learn from vast amounts of data and make autonomous decisions. As AI becomes more integrated into our lives, these systems need protection.

“As with all applications of AI, we can not expect, nor do we want, AI to operate without human oversight. This is especially important in cybersecurity. We must understand the how and the why of its decision-making. If not, AI systems can be manipulated and weaponized.”

Carmen Fontana, IEEE Member

Preserving Integrity

AI models are vulnerable to numerous threats. One common attack involves data poisoning, in which malicious data is introduced into an AI’s training data set, causing it to learn incorrect patterns or behaviors and leading to inaccurate or harmful outputs.

“Watermarking and metadata embedding can help to support the integrity of training data and model outputs. These methods can be cryptographically verified and authenticated, ensuring the trustworthiness and traceability of the data used in AI models.”

Kayne McGladrey, IEEE Senior Member

A Dual-Use Tool

Increasingly, cybercriminals are using generative AI services to improve and refine their tactics.

“Threat actors have used generative AI to research and discover new vulnerabilities, create malicious code, and refine phishing messages. If someone can use it to their advantage in a malicious context, it's no surprise they do so.”

Steven Furnell, IEEE Senior Member

Rise of Deepfakes

The ability of generative AI to mimic human communication and generate realistic content poses significant challenges for cybersecurity, necessitating more advanced and adaptive defense mechanisms to counter these evolving threats.

“AI can analyze vast amounts of data to identify patterns indicative of cyber threats. Its ability to model and understand normal network behavior allows it to detect anomalies more effectively.”

Inderpreet Kaur, IEEE Senior Member

An Automated Defender

The sheer volume and velocity of AI-driven attacks are far beyond what human defenders can handle alone. It's a high-stakes game of AI versus AI, where automated threats require automated defenses.

“The best defense is meeting AI attackers with similarly powerful proactive AI defense mechanisms.”

Eleanor Watson, IEEE Senior Member

Learn more about emerging issues in AI, Quantum and Cybersecurity.

Human Awareness Is Back in Cybersecurity

Cybersecurity pros emphasize awareness training over technical defenses against rising AI phishing scams.

Read Article

Security in VR and the Metaverse

Augmented and virtual reality raise privacy and security risks, including invasive data collection, intrusive marketing and the privacy of bystanders.

Read Article

Spotting the Deepfake

The field of deepfake detection is booming. Learn how machine learning and other technologies are driving it.

Read Article

The Road to Post-Quantum Cryptography

The security community is testing a new class of algorithms designed to withstand powerful quantum computers.

Read Article

Close Navigation