How AI could change the world
By Iyan Velji, Contributor
When it comes to artificial intelligence or AI, many people are only familiar with OpenAI’s massively popular chatbot, ChatGPT, or perhaps Google’s AI chatbot Bard. Despite this, AI’s capabilities can go far beyond suggesting recipes, writing code and finishing university essays. Researchers, companies, and organizations have been utilizing the burgeoning technology to advance medicine, clean up the planet and even explore the cosmos.
One potential major application of artificial intelligence is in cancer research.
Research from May of this year led by the Harvard Medical School and the University of Copenhagen found that an AI algorithm could detect pancreatic cancer up to three years before diagnosis at a rate more accurate than previous population-wide estimates. This could eventually lead to better detection and earlier treatment which could potentially save lives.
In August 2023, a Swedish study found that when AI systems were used alongside human radiologists, they detected breast cancer at a rate 20 per cent higher than humans working alone.
Beyond merely detecting cancer, AI could play a pivotal role in curing the currently incurable disease.
At the University of Toronto, researchers were able to use Google Deep Mind’s AlphaFold AI software to discover a remedy for a common type of liver cancer in just 30 days.
A separate study spearheaded by researchers at the University of California San Francisco and IBM found that AI can help guide T Cells, a type of white blood cells, in seeking out and destroying harmful cancer cells without stopping. In the future, this type of technology could make regenerative medicine, a field focused on the body healing itself, a more viable possibility.
It’s not just cancer that AI can assist in curing.
There is promising research in AI’s ability to detect Alzheimer’s, according to the Massachusetts Alzheimer’s Disease Research Center, researchers from the New York Stem Cell Foundation and Google Research found that AI can distinguish between healthy and unhealthy cells with Parkinson’s disease.
AI has even been applied to analyze joints with arthritis and joints without it, succeeding with a 75% and 82% accuracy rate respectively, according to research published in the journal Frontiers in Medicine.
In addition to saving human lives, artificial intelligence could play an instrumental role in saving the planet.
According to the International Renewable Energy Agency, renewable energy solutions can provide 90 per cent of the carbon dioxide emissions reductions necessary to reach the goals of the Paris Agreement. Due to this, clean energy solutions like solar, wind, and hydropower will be vital for the 175 countries that have signed the agreement, including Canada.
Luckily, AI can significantly enhance the efficiency and effectiveness of renewable energy solutions to reduce carbon dioxide in the atmosphere, help identify ideal sites, forecast weather changes and reduce the likelihood of outages and blackouts.
AI can also be utilized to help improve and optimize the development of new solar technologies, according to researchers at Massachusetts Institute of Technology and Stanford University.
Like with solar power, AI can forecast weather conditions to optimize the amount of energy produced on wind farms. In 2019, Google’s Deep Mind successfully used an AI neural network to predict wind power output 36 hours before generation. This increased the value of the energy generated by roughly 20 per cent and improved the overall predictability of the energy system.
In addition, AI technology can help optimize wind farms by combatting the so-called “wake effect”. When generating wind energy, turbines at the front of a wind farm sometimes cast a “wind shadow” or wake which can slow down the turbines behind them. In 2022, Wind company Vestas partnered with Microsoft to utilize AI in a prototype for “wake steering”, a technique that mitigates the wake effect, optimizing overall energy production on wind farms.
Finally, AI can help optimize Hydropower energy, which uses the natural flow of moving water to generate electricity. Researchers from the University of Calgary were able to apply a type of AI known as ANFIS (Adaptive Neuro-Fuzzy Inference System) to predict river flow up to six days in advance. Similar to weather forecasting with solar and wind energy, better river flow prediction allows hydro turbines to operate at their most efficient levels, maximizing the amount of energy generated per unit of water.
Beyond river flow prediction, AI can also help with the maintenance of hydropower plants. In China, the State Power Investment Corporation (SPIC) employs a fleet of robots and drones daily to monitor its Wuqiangxi hydropower plant and give updates on the state of the facility in real-time. This has reportedly saved the company 10% on maintenance costs.
A final noteworthy domain where AI is becoming increasingly relevant is the field of astronomy.
A study from earlier this year published in a journal of the Royal Astronomical Society found that AI-powered software could sharpen images of phenomena like galaxies, nebulae, and black holes taken from telescopes on Earth.
AI has the potential to play a crucial role in safeguarding the Earth from asteroid threats. In 2022, an AI asteroid discovery algorithm known as HelioLinc3D was able to identify an asteroid expected to approach Earth within a distance of 225,000 kilometers. Researchers plan to use the algorithm to look for thousands of potentially dangerous asteroids at the Vera C. Rubin Observatory.
Since the time of medieval thinkers like Nicholas of Cusa who wrote “We will suppose that in every region there are inhabitants,” humans have been fascinated by the possibility of extraterrestrial life. With the help of AI, finding this life could become a matter of when, not if, as the technology continues to improve.
Research led by the Carnegie Institution For Science found that a machine learning technique was able to detect whether or not a sample originated from a living entity with a 90% success rate. The researchers have dubbed their findings the “holy grail of astrobiology” and say it could be implemented on NASA’s Mars rovers to help detect alien life on the red planet.
Beyond assisting humans with simple observation of the night sky, AI could become instrumental in helping us physically explore the cosmos.
Earlier this year, NASA was able to utilize commercially available Generative AI software to design lightweight spaceship components that study space, devices that examine the Earth’s atmosphere, space telescopes and even Mars sample missions.
AI can also provide emotional companionship to space explorers as they journey to the moon, Mars, and beyond. In 2018, Airbus and IBM partnered with the European Space Agency to create an AI companion named CIMON for astronauts on the International Space Station.
Even though AI has shown great promise in fields as varied as cancer treatment to exploring the cosmos, it remains imperative to properly regulate the technology to ensure that it is used for beneficial purposes.
According to IBM, AI must be developed with three fundamental principles in mind: respect for individuals working with AI, ensuring AI is used for good instead of harm and fair distribution of the technology among all people.
On Aug. 28, 2023, the US Military announced a new initiative called “Replicator” that aims to deploy thousands of AI drone ships, planes, ground vehicles and other weapons in less than two years. The United Nations and the International Committee of the Red Cross have called for world leaders to set clear restrictions on autonomous weapons systems before 2026.
Another frightening use of AI has been in the creation of revenge and child pornography. In August 2023, the top prosecutors in all 50 American states urged Congress to study how AI could be used to exploit children and craft the necessary legislation to guard against it, according to AP News.
Finally, it will be crucial to ensure that artificially intelligent systems do not perpetuate harmful human biases such as racism and sexism. In 2015, Amazon scrapped an automated system for ranking resumes after the company realized it was discriminating against women, according to Reuters. The company now uses a watered-down version of the recruiting engine to perform simple tasks like removing duplicate candidate profiles.
To ensure that AI is utilized ethically in business, governments, and organizations, mechanisms such as AI Ethics Committees should be established to identify and prevent ethical risks of AI solutions, according to Harvard Business Review.
If AI is developed responsibly and with good intentions in mind, it might just prove to be a game changer for humanity, the world and beyond.