PCMag editors select and review products independently. If you buy through affiliate links, we may earn commissions, which help support our testing.
As it does every year, Gartner used part of its annual IT symposium this week to run down its list of top strategic trends for the coming year. This year’s list, presented by Gartner analyst Gene Alvarez, focuses on AI imperatives and risks, new frontiers in computing, and human-machine synergy. It’s more focused on specific technologies than in recent years, but it started with a few AI themes, of course.
Agentic AI
We all have lists of things we need to get done, so “everybody needs an agent,” Alvarez says. He describes agents as goal-driven systems that can complete tasks with or without us. They can make decisions and act using memory, planning, sensing, and tooling. They may even make websites and applications unnecessary.
Alvarez asked attendees to create a list of things they would like agents to do for them. Software vendors are adding agentic capabilities to their packages, and developers are starting to create them. He urged the audience not to let agentic AI become the next generation of shadow IT or to think of it as just another flavor of Robotic Process Automation (RPA).
AI Governance Platforms
Over the next two to four years, we need to implement platforms that create trust through transparency, ensure AI serves everyone equally (such as checking for bias), and pave the way to build ethics in every AI that gets created, Alvarez said.
He predicts a surge in AI regulations, which will require platforms that ensure compliance. Alvarez warned against “ethics washing” by adopting superficial ethical AI practices as a marketing strategy; AI systems will need to be pressure-tested to make sure they are not biased.
“Responsible AI will be as standard as cybersecurity and just as critical,” Alvarez said.
Disinformation Security
Misleading information—such as the bad information posted on social media about how to get help after the recent hurricanes in the Southeast—has become ubiquitous.
Generative AI has given bad actors the technical means to bypass controls and cause significant harm to enterprises. To combat this, Alvarez suggested looking at disinformation security solutions, which he positioned as an emerging category of technologies aimed at discerning trust, assessing truth, and tracking the spread of misinformation.
We’re seeing synthetic media acting to bypass things like biometrics and phishing controls. So we’ll need things like deepfake detection to identify synthetic media, as well as controls to authenticate real users, such as watermarks.
Post-Quantum Cryptography
The next section of Alvarez’s presentation was on new frontiers in computing. Because quantum computing can break every asymmetric encryption, every organization will have to protect itself over the next two to three years. “This could be bigger than Y2K,” he said.
This involves implementing post-quantum cryptography, a new set of algorithms designed to resist attacks from both classical and quantum computers.
He told attendees to beware of “harvest now, decrypt later,” which is already happening. The solution is not a simple patch, so organizations will have to inventory and replace all their encryption. He noted that postquantum cryptography algorithms may impact performance and are generally not budgeted. He suggested companies update or replace hardware and move sensitive data to postquantum cryptography ASAP.
Ambient Invisible Intelligence
We’ve been talking about ambient intelligence at least since RFID tags were first introduced, but cost has held it back. We’re now seeing ultra-low-cost, tiny wireless tags, devices, and sensors. Alvarez expects to see 20-cent tags in the near term, dropping to 10 cents in the next five years.
In the long term, this will result in deeper integration of computing into all aspects of life, creating “smart everything,” he said. You’ll need to prepare your infrastructure for millions more objects being tracked and set it up for privacy (e.g., tracking ice cream cartons when they are on a truck or in a store but not at home).
Energy-Efficient Computing
We’re seeing increased demand for energy, driven by AI workloads and rich media content. As a result, energy consumption is rising in an unsustainable manner, so we’ll need innovative technology to address this, Alvarez said. For the long term, IT executives need to monitor new technologies that promise 10 to 100x improvements for things like AI.
In the short term, we should adopt things like green cloud providers, rewrite some algorithms to save energy, and consider shifting the load of some applications to different times of the day.
Hybrid Computing
New computing paradigms are coming, including neuromorphic, quantum, optical computing, and DNA computing, Alvarez said, noting that these will become more important over the next 10 years. But he said that we never switch completely to a new paradigm, noting that mainframes are still in use decades after the introduction of the client-server model. Instead, we’ll use these new technologies along with classical computing for different tasks. He suggested attendees start looking at integration and orchestration platforms that can support the development and deployment of applications across a hybrid paradigm.
Spatial Computing
The final three trends fit into what Alvarez called human-machine synergy. Pointing to things like the Apple Vision Pro and other mixed reality headsets, Alvarez discussed spatial computing, which he described as a computing paradigm that combines physical and digital objects in a shared frame of reference. This allows users to interact with digital content in a 3D space, making it a more immersive and intuitive experience.
In the next one to three years, Alvarez expects new opportunities (such as a virtual workroom) that enable more inclusivity and offer just-in-time contextual information. Gartner estimates this is a $35 billion market in 2024 and growing.
Polyfunctional Robots
Alvarez said his favorite trend is robots that can do a variety of tasks for humans. Such robots will need to be instructed in what to do but not how to do it. We’re now seeing humanoid or canine-style robots, he said, and by 2030, Gartner predicts that 80% of humans will engage with smart robots every day. These are going to become economically viable in environments where a single application would not provide a sufficient ROI, such as private homes. He said that human-robot relationships (“rob otology”) will be increasingly critical to business success.
Neurological Enhancements
The final trend Alvarez mentioned was enhancements that read into your brain and understand what’s going on in it to enhance its functions, like a device that could reduce Parkinson’s tremors, but he said this is more than 10 years out. There are a variety of approaches, including neuromodulation, cognitive training, neurofeedback, and neuroagents, but what’s important now is to watch for bidirectional brain-machine interfaces integrated into earbuds or other wearables.
