The Internet of Things (IoT) has been around since 1990 — ever since John Romkey created a toaster that could be switched on over the internet. Today, 66% of North American homes have at least one IoT device, such as a smart speaker, bulb or watch.

But for all their conveniences, many IoT devices are limited in functionality and performance. Moreover, they have notable security flaws that could compromise public safety, consumer data or entire company databases.

The key to unlocking the full potential of these devices and improving security in the IoT industry as a whole may lie with another rapidly rising force in modern technology: artificial intelligence.

The rapid evolution of artificial intelligence

IBM defines artificial intelligence as software that “leverages computers and machines to mimic the problem-solving and decision-making capabilities of the human mind.”

Since Alan Turing asked, “Can machines think?” the world has been fixated on finding the answer. Just a decade ago, AI was still relatively unproven in the mainstream. But progress has been exponential, with investment in AI startups increasing sixfold since 2000.

During the COVID-19 pandemic, many organizations adapted to remote work environments by embracing the subfields of AI —  automation, machine learning, data science and deep learning.

Here are three of the most common uses of AI:

• Speech recognition uses natural language processing (NLP) to translate human speech into text. Most modern mobile devices have built-in speech recognition software for voice search and text messaging features.
• Customer service chatbots have replaced many human support representatives. From FAQs to handling complaints, AI-driven messaging bots offer tailored assistance depending on the needs of each customer.
• Recommendation engines on e-commerce stores use AI algorithms to study consumer shopping data trends and offer targeted upsells and cross-sells to enhance the online shopping experience.

In The State of AI in 2022, McKinsey reported that the adoption of AI has more than doubled in the past five years. However, the proportion of organizations using artificial intelligence has plateaued — settling between 50% and 60%. As we move into 2023, we can expect AI to tackle a long-standing problem — IoT security.

The issue with the internet of things

The Internet of Things (IoT) is the collective network of sensors, tools and physical objects that contain software and technology that enable internet connectivity, data exchange and communication with other connected devices and systems.

Some popular IoT devices include:

• Smartwatches let you do many smartphone tasks while you’re on the go, including calls, messages, emails and payments.
• Smart bulbs are easy to install and use. You can control lightning, color, intensity and timers from your smartphone or with voice commands.
• Smart TVs support Wi-Fi, Bluetooth, Ethernet and USB standards. You can connect wirelessly to your smartphone or tablet and also watch content on streaming services.

While IoT has always been an innovative space, there are undeniable drawbacks to the technology. Many products suffer from common issues such as disconnecting or delayed responses. But that’s not the biggest concern.

What are the security problems in IoT?

Because IoT devices transmit large amounts of data, they naturally attract the attention of cyber criminals.

Whether it’s an identity thief snooping for personal information on public Wi-Fi networks or state-sponsored hacking groups looking to breach government entities, IoT devices are vulnerable points in any security posture.

Here are a few of the main security issues with IoT:

• Lack of visibility. Most manufacturers have no way to monitor their devices after they are shipped to users. Without oversight for maintenance and security, a single malfunction could expose the device to a data breach.
• Code vulnerabilities. Developers of IoT devices typically use open-source software, which is prone to bugs and security vulnerabilities.
• Poor security integration. Security is often an afterthought in IoT. In 2021, Forescout reported “at least 100 million IoT devices” were not adequately secured. Many of these vulnerable devices are in key industries, including medical services.
• Weak passwords. Many users don’t change the default passwords on smart home devices, leaving them at risk of being compromised by hackers.

The lack of visibility and testing in IoT is an ongoing issue. Many devices have unpatched vulnerabilities, making it a question of when rather than if they will be hacked.

But now, thanks to the rapid advances in AI technology, there is a solution that could revolutionize security in the IoT industry.

The growth of Artificial Intelligence of Things (AIoT)

Artificial Intelligence of Things (AIoT) combines the data-driven knowledge of artificial intelligence technologies with the connected devices from IoT infrastructure.

AI technology can improve the efficiency of IoT devices through two major subfields:

• Machine Learning (ML) enables computers to rapidly analyze large datasets and continuously learn from them to produce precise outcomes
• Deep Learning (DL) is a subset of ML that trains computers to think like the human mind and solve more complex problems.

With artificial Intelligence, IoT devices can collect and analyze data. Moreover, these devices can then use the learned information to simulate smart behavior, such as making autonomous decisions to switch on a bulb or reduce the heat on a thermostat.

But how can this union of AI and IoT tackle the major security problems in IoT devices? Here are several use cases of AIoT that offer better security for people, data and property.

Better data security

Developers can incorporate AI technologies like decision trees, linear regression and neural networks to create more effective IoT cybersecurity applications that are able to identify and nullify threats quicker.

Improved visibility and analysis

As AI facilitates high-speed analysis with large data sets, we can improve how IoT devices manage and monitor data. Companies that integrate IoT with AI and ML technologies benefit from continuous monitoring and analyses and real-time situational awareness. Ultimately, this approach leads to accurate decision-making with minimal human interference.

More robust access control with smart locks

AIoT can improve access control in multiple industries. For example, banks and healthcare institutions can use AIoT-enabled devices equipped with biometric technology to restrict access to specific rooms, vaults or equipment storage compartments.

Safer workplaces with smart factory robots

AI and IoT have been embedded in manufacturing for years. Now, as the technologies converge, smart factory robots with implanted sensors can facilitate data exchange. Not only will this transformation save time and money, but it will also reduce human error and injuries as manufacturing operations become more automated.

Creating safer self-driving vehicles

AIoT enables autonomous vehicles to analyze weather, road conditions and speed and adjust accordingly. Despite several setbacks, Tesla looks destined to create automobiles that are capable of detecting animal and pedestrian activity and safely navigating journeys without the need for human drivers.

Improving operational efficiency in large workplaces

Digital twins use sensors to collect real-time data about a physical object and create a digital duplicate that analysts can review, optimize and learn from. This use of AIoT can be applied to large retail outlets or industrial environments, allowing managers to analyze the activity of the real environment and respond to changes or incidents quickly.

Improve traffic management in smart cities

Over 50% of smart cities have adopted 5G technology, which makes communication between connected devices faster and more reliable. One example is the rise of AIoT-based drones. These devices transmit and analyze real-time traffic data and make decisions to control traffic lights and reduce congestion.

As smart cities, offices and homes embrace the global rollout of 5G, we can expect AIoT adoption to grow at scale in 2023.

AI and IoT are natural allies that empower each other

Artificial intelligence unlocks more potential in the internet of things by enabling connected devices to analyze and learn from past data, predict future activities and make smarter, data-driven decisions. In return, IoT adds value to AI technology through increased connectivity and data exchange.

While AIoT devices bridge the security gap in IoT, we can get more peace of mind knowing personal and company data is safer with these devices. The growth of AIoT paves the way for more effective, secure, future-proof systems and applications for companies and consumers in 2023.