Adaptability Separates the New AI from the Old One

Adaptability Separates the New AI from the Old One
Adaptability Separates the New AI from the Old One

Artificial intelligence has numerous definitions and has long been a part of software, but current developments can tailor learning, coach students, and automate certain areas of school administration.

New artificial intelligence technologies have been discovered by certain educators to help them in their profession. Some people are hesitant or perplexed. Since the middle of the 1950s, the word “AI” has been used, but recently it has become much more popular. Different perspectives on the future of AI and its effects on the world are held by Elon Musk and Bill Gates. Guardrails and major actors are in the news.

I discussed the moral issues with AI in student essays and college admissions last November, and I recently found that teachers are using AI to create recommendation letters. I have discovered some helpful approaches to convey a few AI fundamentals in this wide and difficult subject in my work with educators who are perplexed about the subject.

From extremely rudimentary to very complex, AI is a continuum. For a very long time, we have had and employed AI. The new AI is distinctive—very distinctive.

AI is defined in a variety of ways. I find it easiest to comprehend AI as using machines (computers) to replicate human intelligence. A benchmark known as the Turing test states that effective AI would make it difficult to determine whether one was speaking with a human or a computer program.

Must read: Artificial Intelligence Becoming a Game Changer for Intel Stock

We’ve had software and hardware that seemed to think for a while. All of the everyday objects we use that have basic artificial intelligence (AI) already built in—like our printers and cars—go unnoticed. The limitation of old AI, or primitive AI, is that it is incapable of learning how to deal with anything novel outside of what its programmer had anticipated. The built-in computer code in my printer allows it to operate and respond.

The designers made an effort to consider every scenario in which a printer can malfunction. They set up detectors to keep an eye on its performance and health. Now, the printer manufacturer sends me a software update if they discover a way to fix or enhance the printer. My printer can only carry out the commands that the programmers give it.

Aha! These machines can’t change their lot in life, so to increase their intelligence, they need to learn from people outside of their field. Although programmers figure out how to make my phone’s camera better, it doesn’t make it any better at taking images, so I get the software upgrade.

You see what I mean. Let’s distinguish between the new AI, which includes learning, problem-solving, and improvement built in, and programmed intelligence. The two types of emerging AI that experts distinguish are limited and general.

Must read: The role of artificial intelligence in ecommerce

v Narrow Ai

Narrow AI only employs its learning in a small range of applications. The ability to trade stocks cannot be used by an AI algorithm to identify illnesses in medical specimens.

v General Ai

Similar to how a human might utilize their general intelligence to apply what they know extensively in life; general AI would learn and grasp things that can be useful in different contexts. It is outside the purview of this essay to discuss Musk and Gates’ hand-wringing about the future because general AI is not yet a reality.

I acknowledge that comparing ancient and modern AI is oversimplified and that the two can coexist, but I believe that viewing it as a continuum might be helpful for parents and educators. We can discover where each new AI component falls on that continuum, which can help soothe worries of the unknowable. It can inform instructors about the tools’ characteristics and aid them in decisions they are making about hardware and software.

Numerous math applications, for instance, can assist pupils in learning addition, subtraction, and other concepts. Not all are excellent or successful. The AI in good programs can be adjusted to a student’s performance to assist them.

Must read: The advancement of science in the era of artificial intelligence

Re-drilling missed facts might help pupils move more quickly when they are quickly “getting it” or more slowly when they miss too many. Due to the programmer’s careful consideration of every possible student response, these programs can behave like human partners. But when a pupil is struggling, the program doesn’t get smarter or perform a better job. The software is updated by the programmers based on outside views and experiences. With time and experience gained from both their successes and failures, new AI math algorithms would get better

We must be conscious that there are drawbacks to all technology, and the use of AI in classrooms raises several issues that cannot be dismissed. Along with prejudice at the core of the program and its algorithms, deep fakes, intellectual property rights, worries about Big Brother, and dangers to humanity and the heart of education, cost and digital equity are two other important considerations. There are a lot of other issues that need to be addressed as well. It’s not the first time a new technology has sparked ethical debate, but this one’s ramifications for the future are novel.

AI is here to stay in this brave new world. As educators, it is crucial that we make every effort to thoroughly comprehend how to use these tools in classrooms and schools for the benefit of our students and to properly instruct kids in their use.

Must read: The new synthesizer — or just muzak — might be produced by Meta’s AI music generator.

Leave a Comment