New AI Can Pass Every Single Test And Exam In The World
ChatGPT-4, the latest version of the power AI software, passed the Bar, LSAT, and SAT with an almost perfect score.
It’s hard to believe that although we’re living in the present day, we’re also witnessing future technology beyond our wildest imagination develop in front of our very eyes. We’re talking, of course, about OpenAI’s ChatGPT-4, a large language model (LLM) powered by artificial intelligence (AI). And according to Futurism, the technology is only improving, citing above-average scores on tests such as the Bar, the LSATS, and the SAT’s reading and math tests.
While the applications of AI in generating the written word are staggering and endless, the technology is still far from perfect. As it stands, ChatGPT-4 is the most sophisticated version of the software, but it still has similar limitations to its predecessors that require human intervention. The issue with AI technology is built right into the name; it’s not fully reliable because its intelligence is artificial.
Sure, ChatGPT-4 can ace tests, but when used for more open-ended applications, we’d strongly recommend proofreading the AI-generated content before popping off any important emails or submitting any book reports to your professor (especially if your final grade depends on it)! Even OpenAI cites “reasoning errors” that require a human touch to overcome, stating that one should avoid high-stakes uses altogether until the technology becomes more reliable. In other words, if you’re an overtired, over-caffeinated college student trying to bang out a report at the last minute, it’s probably still a good idea to know your source material.
Ethically speaking, the use of AI occupies a lot of grey areas. A recent episode of South Park dives into the implications of using apps like ChatGPT to respond to texts or complete schoolwork. And there’s no doubt that some of us wonder what will happen when AI technology learns to reason and develop beyond our comprehension. After all, many science-fiction stories that involve robots overthrowing humanity start with artificial intelligence becoming smarter than its human counterparts.
We know that it’s a huge leap to liken the use of AI and LLMs to facilitate projects and test-taking to robots taking over the world like in The Terminator, but in this case, it would be terrifying if life ultimately ended up imitating art. The flip side is that a lot of busy work can be bypassed by implementing AI technology into the workforce, freeing up time for projects that are more important than popping off emails at the office. The important takeaway is that the current state of AI is fascinating and useful but still has a long way to go before being utilized fully without human intervention.
Amazon, however, is already selling books written by AI, and you could see why writers could be worried about their livelihoods being at stake. But for now, applications like ChatGPT are still imperfect and lack the imagination that their human counterparts have. At the end of the day, we’re simply living with an alternative technology that has proven itself to be useful when put in the right hands. And if you’re worried about your kids getting kicked out of the university after getting caught cheating, there are already applications in place to detect the use of AI.
So if you’re a writer, an artist, and a master of your craft, you’ll likely never have to worry about competing with AI on creative endeavors. But if you’re trying to clear out your inbox so you can get back to work, you may want to give the software a look. If used properly, artificial intelligence is another tool to help us conquer our workloads and get back to what’s important in life!
