NOTES FROM THE FIELD
Technology
Predictions 2025
Dejan Milojicic , Hewlett Packard Enterprise
IEEE Computer Society technology experts have unveiled
› U.S.-centric reduction in
22 breakthrough technologies set to redefine industries interest in sustainability due
to new economic and sociopo-
and shape the future of our world for decades to come. litical pressures (not globally,
F
though)
› ever-increasing automation in
or the past 15 years, in January, the IEEE Computer Soci- many dimensions, setting the stage for additional
ety has been predicting technologies likely to succeed AI opportunities
during the following year. In the past few years, we ›
biotechnology’s rapid development under the radar (for
have formalized this process and expanded it beyond example, “AI-assisted drug discovery” and “AI-based
our fields of interest, embracing, for example, power and en- medical diagnostics”).
ergy, space, health, and other technologies. We also increased
our diversity in terms of global presence, gender, culture, etc. The 22 Technology Predictions for 2025 were broadly
The 54-member 2025 Technology Predictions team made in six categories: verticals (six predictions); applied
foresees the following: AI (four predictions); user interfaces (two predictions); non-
functional characteristics (four predictions); applied com-
› accelerated growth in many artificial intelligence puting (three predictions); and energy-related (three predic-
(AI) facets, requiring reskilling of the workforce tions). The team evaluated technologies for their likelihood
of success in 2025 (see the x-axis in Figure 1), impact on
humanity (see the y-axis in Figure 1), maturity (color coded
This article has supplementary downloadable material available at https://doi. from very early to commercialization in Figure 1), market
org/10.1109/MC.2024.3511020, provided by the authors.
adoption (proportional to the bubble size in Figure 1), and
Digital Object Identifier 10.1109/MC.2024.3511020
adoption horizon (not presented in Figure 1; please see the
Date of current version: 9 January 2025 full report in the supplementary material).
88 CO M PUTE R P U B LISHED BY THE IEEE COMP UTER SOCIE T Y 0018-9162/ 25©2025IEEE
The most likely to succeed and the In addition to multiple insights, we transformation, sustainability, and arti-
largest market adoption in 2025 is also made recommendations to in- ficial general intelligence.
large language model (LLM) deploy- dustry, government, academia, and Like in the past years, we will again
ment; the technology with the most professional organizations, such as develop a scorecard for our 2025 predic-
impact on humanity is AI-assisted IEEE, as well as different roles, such as tions, and we will release it in the last
drug discovery. Some of the insights end users, developers, chief executive/ months of 2025. We actively improve
we made are the following: technology/finance/data/etc. officers, our processes and synergistically de-
and investors. velop them with the FDC’s megatrends.
› Small modular reactors offer
a breakthrough solution for a
sustainable and reliable data center
energy supply. For the first time, the team has identified
› Wearables/biomarkers will key disruptions in 2025 and beyond 2025,
continue innovations in sensor which will influence technology development
accuracy, energy efficiency, and
real-time analytics.
and adoption.
P
› Next-generation cyberwarfare
is emerging as a key area due For the first time, the team has lease read our full report in
to rising concerns around data identified key disruptions in 2025 and t he supplementar y material
integrity and AI vulnerabilities. beyond 2025, which will influence and provide feedback (contact
› The convergence of technologies technology development and adop- i n for mat ion ava i lable at t he end
through systems approaches (for tion. We also mapped all technologies of the report). Also, please consider
example, IT/energy sectors) is to the IEEE Future Directions Com- joining us in developing future
predicted. mittee (FDC) megatrends:1 digital r eports.
Impact on Humanity Higher
Than Chance of Tech Success
(Worth Investing in)
Very Early Prototype Incubating Emerging Mature Commercialization Chance of Success Correlates
to Impact on Humanity
5 Al-Assisted Drug Discovery
A
Tools and Policies for Al Regulation
Wearables/Bioimarkers in
Chance of Tech Success Higher
Al-based Modical Diagnosis Than Impact on Humanity
Medicine/Wellness
Impact On Humanity
4.75
SmartAg
Functional Safety/Autonomous Vehicles
Sustainable Computing
IT/Energy Convergence
A/B
4.5 Al-Enhanced Robotics
New Battery Chemistries Al Agents
4.25
Brain–Computer Interfaces Mis/Disinformation
Augmented Al
B
LLM Deployment
4 Autonomous Driving
Nuclear Powered Data Centers
Al-Optimized Green HPC
Drone Adoption
3.75 Next-Generation Cyberwarfare
Data Feudalism
B/C
Space Computing
3.5
3.25 3.75 4.25 4.75
C+ B/C B– B B+ A/B A–
FIGURE 1. Comparing 2025 Technology Predictions, Clusters of Correlated Technologies. Technologies highlighted in light green
have a higher impact on humanity than the likelihood of success and are worth investing in by governments. Those highlighted in
light red have higher chances of success and are worth considering by the industry for further commercialization. Prediction: Tech.
development in 2025 (x-axis) versus impact to humanity (y-axis) (size of bubble proportional to relative market adoption). HPC:
high-performance computing; LLM: large language model; SmartAg: smart agriculture.
JANUARY 2025 89
NOTES FROM THE FIELD
ACKNOWLEDGMENT Keqiu Li, Avi Mendelson, Cecilia Me- REFERENCE
The authors of the IEEE Computer tra, Dejan Milojicic (chair), Puneet 1. IEEE Future Directions. “2024 tech-
Society Technology Predictions 2025 Mishra, Christine Miyachi, Khaled nology megatrends.” IEEE. Accessed:
include: Ali Abedi, Mohamed Amin, Mokhtar, Chengappa Munjandira, Bob Dec. 10, 2024. [Online]. Available:
Cherif Amirat, Jyotika Athavale, Parro, Sudeep Pasricha, Nita Patel, https://bit.ly/get-megatrends
Mary Baker, Greg Byrd, Kyle Chard, Alexandra Posoldova, Marina Ruggi-
Tom Coughlin, Izzat El Hajj, Paolo Fara- eri, Tomy Sebastian, Farzin Shadpour,
boschi, Rafael Ferreira da Silva, Nicola Sohaib Sheikh, Saurabh Sinha, Vesna
Ferrier, Eitan Frachtenberg, Jean-Luc Sossi, Luka Strezoski, Vladimir Ter- DEJAN MILOJICIC is an HPE Fellow
Gaudiot, Ada Gavrilovska Habl, Alfre- zija, George Thiruvathukal, Michelle and vice president at Hewlett
do Goldman, Mike Ignatowski, Lizy K. Tubb, Gordana Velikic, John Verbon- Packard Labs, Milpitas, CA 95035
John, Vincent Kaabunga, Mrinal Kar- coeur, Irene Pazos Viana, Jeffrey Voas, USA. Contact him at dejan.
vir, Hironori Kasahara, Witold Kins- Rod Waterhouse, Stefano Zanero, Gerd
[email protected].
ner, Danny Lange, Phillip A Laplante, Zellweger, and Ying Zhang.
Computing in Science
& Engineering
The computational and data-centric problems faced
by scientists and engineers transcend disciplines.
There is a need to share knowledge of algorithms,
software, and architectures, and to transmit lessons-
learned to a broad scientific audience. Computing in
Science & Engineering (CiSE) is a cross-disciplinary,
international publication that meets this need
by presenting contributions of high interest and
educational value from a variety of fields, including
physics, biology, chemistry, and astronomy. CiSE
emphasizes innovative applications in cutting-edge
techniques. CiSE publishes peer-reviewed research
articles, as well as departments spanning news and
analyses, topical reviews, tutorials, case studies, and
more.
Read CiSE today! www.computer.org/cise
Digital Object Identifier 10.1109/MC.2024.3516457
90 COMPUTER W W W.CO M P U T E R .O R G /CO M P U T E R
