2024 was a year of unprecedented innovation and global upheaval. As we look ahead, Amy Webb, CEO of the Future Today Institute asks which technologies will reshape our world in 2025?
AI may seem like the obvious answer, but it’s just one of three groundbreaking technologies that will shift the business landscape in the coming year. The other two, advanced sensors and biotechnology, are less visible, though no less important. As geopolitical tensions rise and the global trade landscape shifts, these tiny components will play an outsized role. Expect to see innovations in areas like supply chain tracking, autonomous vehicles and even implantable medical devices, all fuelled by advancements in sensor technology. In biotechnology, researchers developed a 3D bioprinting technique that results in the rapid creation of functional biological issues, achieving speeds 10x faster than before and getting us closer to lab-grown tissues and organs at scale.
Two frontier technologies will see breakthroughs in the coming year. The dream of near-limitless, zero-carbon energy from fusion is within reach. Thanks to recent breakthroughs, fusion machines are on the verge of producing more energy than they consume. Soon after, fusion could power data centres, manufacturing, steel mills and chemical manufacturing, enabling us to move away from fossil fuels. In addition, the long-anticipated emergence of quantum computing from the realm of theoretical possibility to practical reality could soon reach an inflection point. The potential for disruption ranges across industries, from medicine to materials science to finance.
In an era of unrelenting change, leaders must resist the temptation to wait for the dust to settle and instead become astute observers of emerging signals, monitoring for subtle shifts in the competitive landscape that portend disruption. Quantum computing may not be accessible today, but actively engaging in ‘what if’ scenario thinking will challenge assumptions and prepare organisations for a range of potential futures, enhancing strategic resilience. These key themes have already emerged for 2025 and are on the radar.
We have entered a technology supercycle
This will be a period of explosive technological advancement driven by the convergence of multiple emerging technologies. Think of it like a perfect storm of innovation, where breakthroughs in various technologies collide and amplify each other. This creates a period of rapid disruption and transformation, similar in magnitude to the introduction of electricity or the internet, leading to entirely new industries, economic shifts and societal changes.
• Living Intelligence is the next big disruption: AI is shifting the business landscape, with sensors and biotechnology quietly advancing, and along with AI, their convergence underpins a new reality that will shape the future decisions of every leader across industries. I call this new reality ‘living intelligence’, which creates systems that can sense, learn, adapt and evolve. Living intelligence will drive an exponential cycle of innovation, disrupting industries and creating entirely new markets. Leaders who focus solely on AI without understanding its intersections with these other technologies risk missing a wave of disruption already forming.
• PLAMs, CLAMs and GLAMs are the next LLMs: As more sensors surround us, they will capture and transmit not just more data, but more types of data. While organisations are busy creating and using LLMs, they will soon need to build LAMs: large action models. If LLMs predict what to say next, LAMs predict what should be done next, breaking down complex tasks into smaller pieces. Unlike LLMs that primarily generate content, LAMs are optimised for task execution, enabling them to make real-time decisions based on specific commands, which will be enormously helpful in organisations of every size and scope. LAMs use the behavioural data we generate when we use our phones or operate our vehicles, along with a constellation of sensors everywhere, collecting multiple streams of data at once. As LAMs become more embedded in our environments, they will operate seamlessly, often without users’ direct engagement.
What many organisations are failing to imagine is how LAMs will evolve to personal large action models (PLAMs), eventually interacting with different systems, learning from large datasets and adapting to changing business needs. PLAMs will improve digital, virtual and physical experiences by streamlining decision-making, managing tasks, negotiating deals and anticipating our needs based on behavioural data. They won’t need conscious input - these autonomous agents will personalise recommendations, optimise purchases and communicate with other trusted agents, allowing seamless transactions − all while maintaining a user’s privacy and preferences. PLAMs, by definition, have access to all of the user data on personal devices. While people will have PLAMs, corporations will likewise have one or more corporate large action models (CLAMs), and digital-forward governments will have government large action models (GLAMs).
• Weird tech alliances are the new normal: At this year’s AWS re:Invent, Apple made a surprise appearance. Apple’s Senior Director of Machine Learning and AI, Benoit Dupin, shared the stage with Amazon Web Services CEO, Matt Garman to announce that Apple is using Amazon’s proprietary chips. Dupin said that Apple is in the early stages of using Trainium 2, the latest version of Amazon’s AI training chip. This highlights Apple’s strategic alignment with AWS as the cloud giant continues to compete with Microsoft Azure and OpenAI. Meanwhile, Amazon and Anthropic are collaborating on Rainier, a massive, distributed AI compute cluster with hundreds of thousands of chips designed to train next-generation models. The future of AI is tethered to the future of the cloud and compute, and we expect to see more permutations involving Azure, AWS and Google Cloud in the coming year.
• The crypto winter thaws: Last year at this time we said the crypto winter would begin in earnest, as it had been a tumultuous year for the crypto industry. While crypto is still volatile, key developments late in the year indicate that winter may thaw early. Bitcoin hit $100 000 for the first time ever, which is a remarkable milestone considering all the turbulence over the past 12 months. The surge is mainly due to Donald Trump’s election and his promise to make the US “the crypto capital of the planet” through deregulation. With a Republican supermajority, more favourable regulations will likely go into play.
• Quantum might finally have its moment: The next couple of years could be a turning point for quantum computing due to several converging factors. First, advancements in error correction techniques are making quantum computers more stable and reliable, paving the way for practical applications. AlphaQubit is an AI-based decoder from Google DeepMind and Quantum AI, and it identifies errors with state-of-the-art accuracy. Accurately identifying errors is a critical step towards making quantum computers capable of performing long computations at scale, opening the doors to scientific breakthroughs and many new areas of discovery.
Second, companies like IBM are actively working to increase the number of qubits in their quantum processors, allowing for more complex calculations. Google’s Sycamore processor has demonstrated the potential for quantum supremacy, while IonQ is developing trapped-ion quantum computers for commercial use. Third, increased investment from both the public and private sectors is accelerating research and development. What we will see more of in 2025 is quantum hybrid systems, which can unlock immediate value for business. Companies like Rigetti Computing are focusing on building hybrid quantum-classical systems that can be integrated with existing infrastructure, making the technology more accessible.
• Extreme weather and climate events will catalyse tech innovation: The escalating effects of climate change will act as a powerful catalyst for tech innovation – and commercialisation – driving the adoption of solutions that had been held up by cost and market demand. Rising sea levels and extreme weather events will create an urgent need for resilient infrastructure, leading to the wider use of technologies like CarbonCure’s concrete additives that reduce carbon emissions during production.
We already have a solution for water scarcity in the form of desalination. Making clean water more accessible will become a priority, and new research from MIT proves that it’s possible to deploy solar-powered desalination systems that don’t require extra batteries. There are new distress sensors for plants in the works, along with synthetic biology-inspired seed coatings and engineered microbes that help decrease carbon emissions.
• Nuclear energy makes a comeback: Many of us vividly remember the Chernobyl and Fukushima accidents, highlighting the dangers of a catastrophic meltdowns and the aftermath of radiation leaks. With new evidence that Russia may have invented a space nuke, there are concerns that nuclear energy projects could be repurposed for military purposes. It may come as a surprise to learn that the tech titans are increasingly investing in mini-nuclear plants, known as Small Modular Reactors (SMRs), as a key component of their future energy strategies. AI’s complex algorithms and massive datasets require a lot of computational power – and within a few years, they will exceed what today’s power plants are capable of producing. Enter SMRs, which offer a more scalable and flexible alternative to traditional nuclear power plants, making them attractive for decarbonising data centres and other energy-intensive operations.
Microsoft has partnered with X-energy to deploy an SMR at one of its data centres by 2028. It hopes to show how this technology can power large-scale operations with carbon-free electricity. Google has committed to purchasing 500 MW of power from Kairos Power’s SMRs to provide clean energy for its data centres starting in the 2030s. Amazon also invested in X-energy and is partnering with utilities to develop SMR projects in Virginia and Washington state. Even before the current AI rush, Bill Gates co-founded TerraPower in 2006, with the aim to develop advanced nuclear reactors, ultimately addressing climate change. TerraPower has broken ground on its first nuclear power plant, the Natrium reactor, near a retired coal plant.
• Chaos in Europe: The political turmoil in France and Germany, with leadership contests creating deep uncertainty, could significantly hinder Europe’s innovation potential in 2025. As these governments grapple with internal challenges, economic instability and policy paralysis are likely to arise. Rather than strategic planning for the future, these governments will probably focus in on the present. This uncertainty comes as the EU’s AI Act demands stronger government oversight and the regulation of artificial intelligence technologies, potentially creating further complexities for businesses, hindering the development and adoption of AI solutions.
• Washington’s Game of Thrones: Like the rival houses in Westeros, tech billionaires and politicians will use their resources – wealth, technology or legislative authority – to outmaneuvre each other at the expense of everyday people. In 2025, the balance of power in Washington could shift dramatically as tech executives, bolstered by immense wealth and influence within the Trump administration, increasingly assert their voices in policymaking. ‘Big tech’ could win the iron throne, kicking ‘big government’ out of the realm.
As we move along in 2025, we remain energised by the opportunity to help leaders and organisations thrive in an era of unprecedented change. Looking ahead, we are committed to building on this momentum to deliver impact at scale. Here’s to yesterday’s lessons, today’s triumphs and tomorrow’s possibilities.
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version