Home Artificial Intelligence Scary ‘Auto GPT’

Scary ‘Auto GPT’

0
Scary ‘Auto GPT’

(Video production = AI Times)

Many individuals are surprised on the speed of development of artificial intelligence (AI) lately, but recently, ‘Auto GPT’ can also be surprising us.

This tool is an AI chatbot that works based on GPT-4 developed by OpenAI. It’s an idea model through which the ‘autonomous iterations’ function that corrects mistakes by itself and separate memory and multifunctionality are added to ‘ChatGPT’. It’s released as open source and might be run on a pc.

In the event you set a goal, it learns and finds a technique to achieve it with out a person providing you with prompt instructions.

It’s as if I, the user, designate an agent to offer work instructions in a broad framework after which let the remaining care for itself. This agent gets the job done by searching the Web, reasoning, and self-learning to search out tools to unravel problems.

Nonetheless, the response of many users is that they don’t just do the job, but do it well enough to feel scary.

A pc engineer named Varun Maya recently shared the strategy of using AutoGPT on Twitter, which became popular. After looking for , I actually created a node to create the server.”

A node is a software platform used to develop network applications.

(Photo = shutterstock)

One other business scholar asked ‘are you able to make me wealthy’ to check this chatbot. Then, AutoGPT opened an online browser on his computer, found the bank site, and tried to attach with the saved ID and password.

He should have been attempting to work out the condition of the property. A startled user swiftly interrupted the operation.

Next, let’s analyze the profit model of the ten largest corporations in the US. He said that he created an excellent article in regards to the profit structure of Apple, Alphabet, and Tesla by reading the entire Wikipedia contents in addition to the references attached to related articles.

Sharing this experience on Facebook, this scholar evaluated the looks of AutoGPT as “a serious event that may change our lives.”

Auto GPT use cases have recently been uploaded to YouTube, and all of them have reactions reminiscent of ‘surprising’, ‘terrifying’ and ‘that is the longer term of AI’. Some say, “Cheating using ChatGPT seems like a trivial thing,” and “warnings against AI are gaining strength.”

AutoGPT was created by British developer Thoran Bruce Richards and released at the tip of last month, and is accessible for download on Github. The user’s computer should have version 3.8 or higher of the programming language ‘Python’ installed, and they have to apply for the API of Pinecone, an open AI and artificial intelligence distribution platform, to acquire a key.

Also, since GPT-4 is used, if the quantity of labor increases, you could have to pay API usage fees to OpenAI.

Next, we’ll let you know in regards to the technology trends.

Nvidia's H100 chip (Photo = Nvidia)
Nvidia’s H100 chip (Photo = Nvidia)

■ The worth of Nvidia’s supercomputer GPU module H100 has risen by as much as 10,000 dollars (roughly 13 million won) in a single 12 months from 36,000 dollars last 12 months to 46,000 dollars recently. As generative AI spreads, so does the necessity for computing power, and in consequence, demand for Nvidia’s chips is increasing, and costs have risen.

■ ‘Persona’ services that allow AI chatbots to have virtual personalities are emerging one after one other. Talking to Shakespeare or Cleopatra, for instance. Depending on the characteristics of the persona, the best way of speaking is different, and it’s interpreted as a trend to distinguish as AI chatbot services are being released rather a lot.

(Photo = shutterstock)
(Photo = shutterstock)

■ Google will launch a latest interactive search engine that mixes search and chat next month. It is a latest engine named ‘MAGI’, and Google initially took the position that chatbots weren’t suitable for search. Nonetheless, Samsung Electronics was surprised to see a move to vary the fundamental search of smartphones to Microsoft, and eventually modified its position to introduce a technique of mixing chat and search like ‘Bing’.

■ Stability AI has released a beta version of ‘Stable Diffusion XL (SDXL)’, a more sophisticated image creation tool. Improved handling of components inside images, reminiscent of hands, and might generate images with simpler prompts.

(Photo = shutterstock)
(Photo = shutterstock)

■ Jobs have been reduced due to AI. Blue Focus, China’s largest marketing agency, said it might indefinitely suspend hiring of copywriters and designers, saying it might introduce AI tools reminiscent of ChatGPT.

■ A famous photo contest winner refused to simply accept the award, claiming that his work was created by AI. The fundamental character is German photographer Boris L. Daxen, who has announced that she’s going to not accept the award for creative on the Sony World Photography Awards.

(Photo = shutterstock)
(Photo = shutterstock)

■ There are cases where AI chatbots reminiscent of ChatGPT provide answers with prohibited content when requested in the shape of a task play. In some cases, chatbots were asked to play the role of grandmothers to get explanations on the way to make bombs, and Linux malware was created.

■ The federal government has decided to support the event and advancement of super-large AI within the private sector. As well as, we decided to introduce a super-giant AI exclusively for the federal government. The federal government announced a plan to strengthen the competitiveness of super-giant AI on the ‘Digital Platform Government Realization Plan Reporting Session’.

Jeong Byeong-il, member jbi@aitimes.com

LEAVE A REPLY

Please enter your comment!
Please enter your name here