A remarkable AI can write like humans — but with no understanding of what it’s saying.
Download MP3
OpenAI’s team reported that GPT-3 was so good that people found it hard to distinguish its news stories from prose written by humans. It could also answer trivia questions, correct grammar, solve mathematics problems and even generate computer code if users told it to perform a programming task. Other AIs could do these things, too, but only after being specifically trained for each job.
Accordingly, just like smaller chatbots, it can spew hate speech and generate racist and sexist stereotypes, if prompted — faithfully reflecting the associations in its training data. It will sometimes give nonsensical answers or outright dangerous replies.
A neural network’s size — and therefore its power — is roughly measured by how many parameters it has. These numbers define the strengths of the connections between neurons. More neurons and more connections means more parameters; GPT-3 has 175 billion. The next-largest language model of its kind has 17 billion .
Entering different prompts can elicit results of varying quality, noted one programmer who blogs under the pseudonym Gwern Branwen. “‘Prompt programming’ is less like regular programming,” he, “than it is like coaching a superintelligent cat into learning a new trick: you can ask it, and it will do the trick perfectly sometimes, which makes it all the more frustrating when it rolls over to lick its butt instead — you know the problem is not that it can’t but that it won’t.
Deutschland Neuesten Nachrichten, Deutschland Schlagzeilen
Similar News:Sie können auch ähnliche Nachrichten wie diese lesen, die wir aus anderen Nachrichtenquellen gesammelt haben.
U.Today - IT, AI and Fintech Daily News for You Today$SHIB 2022 burn roadmap released, Elon Musk says $DOGE is better than anything else, SundaeSwap completes its audit: crypto news digest by Dogecoin ShibaInu Cardano $ADA $XRP
Weiterlesen »
Chinese AI Giant SenseTime’s IPO Comes at a Bad TimeHeard on the Street: China’s SenseTime has overcome headwinds to go public. Its next challenge is to find a path to profit.
Weiterlesen »
What Happens When an AI Knows How You Feel?Technology used to only deliver our messages. Now it wants to write them for us by understanding our emotions.
Weiterlesen »
Chinese Scientists Created an AI 'Prosecutor' That Can Press ChargesResearchers in China have created an AI 'prosecutor' that can reportedly identify crimes and file charges against criminals.
Weiterlesen »
Your Resident McAfee Engineer on Cybersecurity, AI, and Career Growth | Hacker Noon'With all things becoming remote and online, there's a growing risk of cyberattacks like Colonial Pipeline - which could have major impacts on society as a whole', thinks Noonies2021 nominee iamrohitjmathew.
Weiterlesen »
Alexa Shockingly Tells 10-Year-Old Girl To Put A Penny Into An Electrical Socket, Which Provides Important Lessons For AI Self-Driving CarsA recent news report about Alexa telling a young girl to do a dangerous stunt is a reminder that we need to be wary of the things that AI tells us to do, including in the home and even while riding in AI-based self-driving cars.
Weiterlesen »