Ready or not, here it comes — on your Facebook walls and Google searches, in the papers students hand in and the spacecraft NASA send up.

Artificial intelligence is everywhere and it’s only going to become more prevalent in the years to come, say those in the know.

Whether that’s a welcome development or a worrisome one depends on the field, the form it takes and the forecaster, learned in asking University of Illinois faculty, alums and other special guests to predict the future impact of AI in their area of expertise.

“I expect that within five years we will all have the option of a smart, always available, personal assistant.

“A good model would be J.A.R.V.I.S. from the ‘Iron Man’ movie. This assistant will participate in our physical world, likely through AI-powered wearables, such as smart, augmented glasses, while also serving as our guide in the digital world.

“For example, this assistant might remind you of people’s names and relevant background information, help you with directions, simplify ordering items of interest as you move about the world.

“Likewise, it could research topics of interest, help plan your day, prepare answers to emails or inquiries, and optimize your activities around your personal needs.

“While this will provide many benefits, there are also downsides that society should consider, such as personal privacy and agency, increased power of big technology firms and potential risks to mental health.”

“AI will gradually automate much of battlefield functions the way smart home gadgets like smart thermostats, smart lighting systems and smart robotic vacuums automate many of today’s household chores.

“The driving force behind AI’s proliferation in the battlefield will be fear. Fear that if we do not utilize AI first, we shall lose to those who do. AI can be faster than humans and can control operations at larger scales, far beyond the human ability to keep up.

“We might not be comfortable delegating decisions to AI, but we’ll argue that what choice do we have if we are to outpace an adversary whose military machine operates at AI speeds? We might even find comfort in telling ourselves that AI and autonomy will help put humans out of harm’s way. After all, why send people to a dangerous place to do a job that a robot or drone can?

“In principle, AI can also be more accurate. It can help execute operations with surgical precision and reduce unnecessary damage.

“It does not get tired. It does not need food and medications. It is disposable. If captured, it can self-destruct. It’s very convenient.

“In a world where we might entrust our physical safety to autonomous cars and our life advice to chatbots, why not entrust our weapons to robots? It remains to be seen, of course, if AI is indeed the safer shepherd of our guns. We can only hope that the price of lessons learned from that experience will not be prohibitive.”

“The Department of Defense is pursuing artificial intelligence primarily to adopt autonomy in a wide variety of applications.

"These applications will come in two broad forms: autonomy at rest and autonomy in motion.

“Autonomy at rest covers software that enables better human-machine collaboration, such as decision support tools to help humans make faster, more relevant decisions; analytical tools to help analysts to make sense of huge amounts of data; and predictive maintenance and logistics.

“Autonomy in motion will improve human-machine combat teaming through control systems for a wide variety of intelligent, uncrewed systems and guided weapons.

“As more and more autonomous systems are adopted, future warfare will be increasingly defined by humans and machines working seamlessly together to perform all manner of combat tasks, making warfare faster and more lethal.”

“I don’t think the impact on sports will be that dramatic as it applies to fans. Other than using AI to create entertainment for in-game and broadcast experiences, the game will be the game.

“The owners won’t have the guts to make any changes related to officiating. So it will be business as usual there.

“Where AI will have a significant impact is on the medical side. It will touch everything related to player health — proactively and reactively.”

“Looking ahead in the next three to five years, I expect we’ll see AI function as what I call the ‘cognitive microscope’ of education: making learning processes visible in ways that were previously impossible.

“Just as a microscope reveals cellular structures we can’t see with the naked eye, AI will help both educators and learners observe thinking patterns, learning gaps and knowledge connections that have always been there but hidden.

“Here’s what I think this will look like practically: Educators will have AI tools that can instantly identify when a student is struggling with a concept — not by waiting for a failed test, but by analyzing patterns in how they approach problems or ask questions.

“Meanwhile, students will get the kind of immediate, personalized feedback that human educators have always wanted to provide but rarely had time for. Think of it as every student having access to a patient tutor available 24/7.

“The key shift won’t be AI replacing human connection in education, but rather freeing educators to focus on what they do best: mentoring, inspiring curiosity and providing the emotional support that no algorithm can replicate. When AI handles the routine feedback and content delivery, human educators can spend more time on the uniquely human and social aspects of learning.

“This transformation will require us to rethink everything about learning, including what we’re actually measuring, moving beyond tests that AI can easily complete toward assessments that capture genuine human capabilities like creative problem-solving and collaborative thinking.”

“My optimistic prediction is that AI will become a ‘tool to think with’ for students and teachers across all levels of education.

“Rather than using AI to make teaching decisions or create products for students, my hope is that we will use AI to generate rich and realistic raw materials that students and teachers alike can use to enhance problem solving and design.

“For example, a science teacher might ask AI to create an instructional diagram or an animation based on some text that they’ve generated. Or a student in a history class could be given an essay that is created by AI and their assignment is to find errors, or to put the essay in their own voice.

“While I am deeply skeptical of replacing teacher and student work with AI, I think there are many possibilities for augmenting their work by asking AI to find and generate resources that can help us be more efficient, creative and inclusive.”

“AI in space exploration often invokes rogue control systems, like HAL 9000 from ‘2001: A Space Odyssey,’ but large language models (LLMs) have been used at NASA for many years.

“Especially in the space science community, where we are collecting vast amounts of data more than a human can analyze, these tools save significant time in data processing.

“A telescope, for example, can take thousands of images of the night sky, which can be fed into an AI model and when programmed correctly, the AI will pull out the handful of important images that may show a new star or highlight a unique occurrence. Then a human scientist can review those images in detail to make a hypothesis or conclusion — saving them from having to look at all the inconclusive images first.

“As far as human space exploration goes, we are already working on onboard AI models that can assist a human crew in similarly analyzing large amount of data.

“One specific area I have worked on is with vehicle health monitoring that includes a series of sensors that are mounted on the spacecraft that measure everything from temperature, pressure and structural strain that can tell if the vehicle is behaving properly or needs repair. Current systems like this on the International Space Station, rely on those large data sets to be down linked to Earth and a ground computer processes it with human help.

“For deep space missions to Mars, however, we would need that analysis to be done on board the vehicle with fast response to direct the crew if their safety was in danger. These AI models would be able to save crew time and provide redundance safety systems to keep spacecraft healthy for the life of the mission.

“As long as we keep HAL happy, everyone will be happy.”

“There are a lot of misconceptions about what AI can do or will do — perhaps the biggest one is that it will displace many people from many jobs.

“After all, AI can recognize objects in images, answer questions, write computer code and more. But these are narrowly defined tasks and each one requires a dedicated model trained to do just that — and nothing else. Even the most advanced foundation models, like large language models (LLMs), have a limited range of ‘abilities.’

“At their core, LLMs simply predict sequences of words that make sense in a given context. Yes, they’ve been trained on vast amounts of human knowledge, but they’re still just tools. That’s how we should think about AI today: as a toolbox for doing small things faster — the next level of automation for tasks we didn’t know how to automate just a few years ago.

“This shift won’t replace us. It will make us more productive, freeing us from mundane work and giving us more time to focus on what really matters. We’ll still be in control — deciding what to keep, what to discard and how to shape the outcomes. We will have the final say, and we will be responsible for the decisions made even with the help of AI.

“We’ve been here many times before. Every time we invent something that improves our ability to get things done, we move on to the next, bigger, more important thing.

“Some professions change, some fade away — when was the last time you spoke to a telephone switchboard operator, an essential job for over half a century? — and new professions and new jobs emerge.”

“News agencies will emphasize the role of humans in their fact-checking process. The rise of AI-generated content has created a sharp decline in public trust of all news sources because it’s becoming more difficult to figure out what’s real and what’s deep-faked.

“I think we’ll see a growing demand from the public for transparency and human-led investigation to prove there’s a person behind the byline.”

“First, it already is showing amazing skill in predicting day-to-day weather and may extend the ability to predict beyond the traditionally accepted limit of about two weeks.

“Work is underway to apply AI/machine learning models to the prediction of more local weather such as thunderstorms, and the results are promising but much more needs to be done. Also, AI/ML can help improve the process by which we bring observations into forecast models — something we call data assimilation.

“All that said, AI/ML also will be transformative in how we communicate weather forecasts and especially high-threat information like tornadoes in ways specific to a given individual, organization and industry, like electric power generation.

“The key for all users of AI/ML is to understand why it is producing a specific result and the trust one can place in the AI/ML algorithms and the results they produce so that it is not used as a black box.”

“In the era of generative AI, the spread of misinformation poses an unprecedented challenge due to two key factors.

“First, generative AI has drastically reduced the cost of producing misinformation — not only in terms of economic resources but also in the time required to create it. What once took coordinated effort, technical expertise and significant time can now be accomplished in seconds at virtually no cost.

“Second, generative AI systems are remarkably capable of crafting high-quality misinformation that appears credible and persuasive. The sophistication of the language, visuals or audio they produce can make it extremely difficult for the average person to distinguish between authentic information and falsehoods.

“Exactly because of these reasons, curbing misinformation will become a major challenge. Traditional detection and moderation mechanisms — e.g., professional fact-checking — may no longer be sufficient, and both technological and societal solutions — e.g., crowdchecking systems like Community Notes — will be urgently needed to address the scale and subtlety of AI-generated misinformation.”

“In the near term, I think AI will be enormously beneficial when it comes to automated officiating around certain objective calls to maintain the flow and pace of the game.

“I also believe AI will fundamentally change the way fans consume NBA basketball by creating unlimited ways to personalize and customize the viewing experience.”

“AI is a hugely helpful tool that we anticipate will be utilized more and more in the years to come; however, it’s really only as good as what it’s told to do.

“For the tax preparation industry in particular, it may be able to help a firm offload some of the simple or low-level tasks but will still require heavy oversight by tax preparers to ensure the information is accurate and complete.

“When it comes to any kind of complex or nuanced conversation, like in tax planning, the human element is still key.”

“I wouldn’t be surprised if we see AI being used extensively to analyze vast amounts of data, helping law enforcement agencies predict and prevent crimes more effectively.

“This could involve identifying patterns and trends that human analysts might miss, assisting in real-time decision-making during critical incidents.

“Of course, it’s essential to consider the ethical implications and ensure that these advancements are implemented responsibly, with transparency and accountability.”

“AI is being used in many ways across the advertising industry — in targeting audiences more precisely, in automated media placement, even in content creation, with Coca-Cola among the first to use AI. However, humans are still needed to develop consumer insights, strategy and genuinely novel ideas.

“In political advertising, there is some extra regulation — at state levels and in discussion at the federal level — not for outright banning of AI but for disclosure of the use of AI in the content.

“I suspect AI will continue to be used to help political strategists in creating, testing and managing campaigns. This means we’ll probably receive even more personalized ads — I’m hopeful that we, as citizens, are able to know what is AI and what is not, and what is truthful and what is not, so we can make more informed decisions in the voting booth.”

“I liken it to artificial flowers. I remember plastic flowers from my childhood. They had obvious tells and giveaways, and were often made from a cheap plastic that faded in weeks.

“Artificial flowers now are vastly better. They imitate a much wider variety of flowers, and much more successfully. They are well-made, and keep their colors for months or years even if left in the sun.

“They’re still artificial flowers.”

“I wonder if AI might be the tool to help teams do a better job of cracking the code of injuries, especially to pitchers. It will be used in a wide variety of ways in baseball — and all sports — but if it could allow clubs to better predict and prevent injuries, that would be a game-changer.”

“Given the rise of AI and the advent of online collaboration opportunities — e.g., social media, crowdsourcing — emerging research has started to investigate the integration of AI and human intelligence, especially in a collaborative social context.

“This creates unprecedented challenges and opportunities in the field of social intelligence, where the goal is to explore the collective intelligence of both humans and machines by understanding their complementary strengths and interactions in the social space.

“In a nutshell, I don’t think the future relationship between humans and AI will be a zero-one sum game. Instead, I believe humans and AI will work together to explore their ‘collective intelligence’ to make our world a better place.

“Such collective intelligence will be a more powerful paradigm than human intelligence or AI alone and address pressing challenges across many critical industries and applications domains, including health care, disaster management, finance, the environment and smart cities.”

“Current Internet protocols developed in the 1970s and ‘80s will be augmented and gradually replaced over the next 20 years by new, more efficient ones designed for AI-to-AI and human-AI interactions. This means the whole architecture of today’s internet will eventually be replaced.

“Entertainment has always been the primary driver behind changes in the internet’s underlying architecture. It forced its evolution from connecting academic computing centers to serving mass audiences. Video streaming led to the development of several new protocols. As augmented reality (AR), virtual reality (VR) and real-time AI interactions become more prevalent, existing internet protocols can’t handle them effectively, which is pushing the development of new protocols.

“While the protocols themselves are neutral, they enable very different user experiences. On the positive side, these new underlying protocols will enable internet applications to adapt in real-time to user behavior and serve up highly personalized content.

“On the negative side, applications like VR enable richer communication of expressions but use real-time biometrics like eye and heart rate tracking and assessment of your emotional state from facial expressions.

“The combination of this data with the ability to individually tailor content creates a real possibility of psychological persuasion by bad actors.”