If you are wondering what the future looks like, it’s called ChatGPT. Everyone is talking about the artificial intelligence (AI) program that generates text responses to complex requests in just seconds and carries on conversations in a humanlike manner. ChatGPT has dazzled the world with its creativity, flexibility and speed. It can compose original songs, plan your vacations and write computer code. It recently took the bar exam and scored in the 90th percentile.
But ChatGPT also has stoked fear over how we will be affected. President Biden called on Congress to pass bipartisan legislation to limit how much personal data tech companies can collect and to prioritize consumers’ health and safety in product development.
Giant tech companies are spending billions to incorporate chatbots into services we use every day from Gmail to Microsoft Word. Whether you like it or not, says leading technology futurist Martin Ford, AI will transform nearly everything we do, much like the Internet and smartphones have. Bottom Line Personal spoke to Ford about why he thinks AI is so transformational…how it will improve your life…and the potential dangers.
Q: What is “artificial intelligence” actually?
AI is software and machines that think with humanlike intelligence. In the past, computer algorithms followed a series of instructions to complete a task. Example: A thermostat has sensors to detect temperature and instructions to turn on the furnace. AI analyzes your patterns and makes predictions and decisions. It continuously improves with more data. You’ve been using AI for years in your home and office. Examples: Netflix analyzes your viewing history to recommend shows…facial-recognition AI allows you to unlock your smartphone by scanning your facial biometrics…banks and credit card companies use AI to identify unusual spending or withdrawals from strange locations…AI digital assistants such as Siri help you navigate the Internet.
Q: Is it like a supercharged personal assistant?
Chatbots like ChatGPT, Google’s Bard and Baidu’s ERNIE are rock stars of AI for good reason. Rather than just fetching websites or information, they search through billions of books, articles, websites and posts and use that data to respond to requests. Best of all, they communicate in natural, conversational text. Example: Say you want to write a complaint letter to a company. You can go to ChatGPT, explain in a sentence or two what you want, and in seconds, a complaint letter is written for you! Even better: You can go back to ChatGPT and tweak the response by saying “Make the letter angrier” or “Make the letter sound like a lawyer wrote it.”
Q: AI has been around for years. Why is it making this quantum leap now?
Three technological events have coalesced. First, computers got much, much faster. AI uses special graphic microprocessing chips (GPUs) that were developed to render high-resolution graphics in videogame consoles. There’s a new supercomputer optimized for AI that can perform a quintillion calculations per second (that’s a billion billion).
Second, powerful mathematical systems were created based on the way the human brain processes information. These “neural networks” learn skills by finding statistical patterns in enormous amounts of data. Scientists also use Large Language Models that figure out the probability of which words to use in any given context based on the vast trove of Internet texts that the neural networks have trained on. These computer algorithms are so fast that they almost appear to have human thought.
Finally, the digitalization of nearly every aspect of our lives created mountains of data critical for the algorithms to learn from. Ninety percent of all the world’s data was generated over the last two years.
Q: Besides chatbots, how is AI going to affect us in the future?
In the near-term, the ability of software to digest and interpret volumes of data has plenty of applications, especially for our health and aging. Examples…
Detecting cancers that doctors miss. According to The New York Times, Hungary has one of the largest AI screening programs for breast cancer imaging tests. At one Hungarian clinic using AI screening, the cancer-detection rate increased by 13% because more malignancies were identified. Similar techniques are being used in Britain to predict strokes and heart attacks.
Identifying early-stage dementia and Parkinson’s disease. According to a study in Japan, nearly 9,000 individuals were asked to read short sentences aloud. An AI system compared speech to distinguish between healthy individuals and those with subtle cognitive impairment. Another study recorded the cursor movements of elderly people using a computer mouse. AI algorithms were able to detect Parkinson’s disease by comparing the results to how cognitively sound individuals used a mouse.
Q: What dangers does AI pose?
It’s hard to predict what new skills the algorithms will learn or how the public will use those new skills. Advancements in AI will constantly challenge our culture and institutions to regulate and control it. Dangers to be aware of…
Deep fakes. AI-generated audio technology is becoming so realistic that it’s possible to clone voices of people from the audio content they post online.
Career displacement. AI chatbots will make some workers such as doctors and lawyers more productive. Those same chatbots eventually will replace people whose work is repetitive or formulaic. According to a study at University of Pennsylvania, the jobs most at risk are information-processing roles such as public-relations specialists, news reporters and financial analysts. Jobs least affected: Short-order cooks, motorcycle mechanics, and oil-and-gas field workers.
Institutions letting AI make decisions about your life and health. Medicare Advantage plans are increasingly using AI to determine if your care is approved for coverage…and the AI programs are driving coverage denials to new highs.