Chat AIs Friends or Foes – a Very Topical Issue
What are Chat AIs?
Chat AIs – friends or foes
“Chat AIs are computer programs that are designed to communicate with humans in natural language, usually through text-based interfaces such as messaging apps or websites. These programs use various techniques such as machine learning and natural language processing to understand and generate responses to user input. Chat AIs are used for a variety of purposes, such as customer service, information retrieval, and entertainment. They are becoming increasingly popular as they can provide quick and convenient support to users 24/7 without the need for human intervention.“
I wanted to know more about Chat Apps which use Artificial Intelligence. I asked one of the most recently released apps ChatGPT for a simple explanation of Chat AIs and the opening paragraph of this blog was the answer.
We are already familiar with Chat Apps. The Google Assistant can answer mundane questions about recipes in the kitchen and play music on request. Siri can answer your queries once you get over the embarrassment of talking to a machine.
Moreover, Chat AIs are useful in customer service, language learning, and personal assistance. It may be annoying to have to talk to an automated voice but it does save time when asking queries of a technical nature. Having just written that I tried to contact MyAgedCare on behalf of a relative. An automated voice told me the wait time was thirty two minutes so not sure about the saving time bit.
Chat AIs Friends to Some and Foes to Others
Specifically at present there is a lot of discussion about ChatGPT and TikTok. In the last week another European Government asked their employees not to use TikTok for work related matters. Apparently there is unease because TikTok is a Chinese company. It uses natural language processing, Artificial Intelligence and algorithms. As an Ai language model it can find out information on its users. The Chinese Government could maybe access sensitive data.
Additionally, on the TV news this last weekend, I heard that ChatGPT can produce fake emails. This is because it is able to mimic the writing style of any individual. Today 21st April I read that Michael Shumacher’s family are suing a publication which claimed to have an exclusive interview with the former World Champion. It was an AI app that produced the “personal interview”.
ChatGPT is the fastest growing app on the World Wide Web. Earlier this month I heard the Italian Government has banned Chat GPT and the Irish Government was possibly going to do the same. What are Chat AIs strengths and weaknesses and why are they considered a threat?
Are they really so dangerous? Last week Elon Musk and a group of like-minded individuals argued against AI language model developments. Proper controls should govern how they develop. Why are Chat AI’s a problem. Stephen Hawking warned about the development of Artificial Intelligence systems being a threat to humankind. Hence we will look at ChatGTP “the new kid on the block” to see why there are concerns.
ChatGPT
What does ChatGPT stand for?
I had read that ChatGTP (not the snappiest of names) was a very popular app. It could answer questions both simple and complex using natural language. It is easy to download the app so I did and asked it the following question and received the following answer.
ChatGPT stands for “Chat Generative Pre-trained Transformer.” The “Chat” part refers to its ability to have conversations with people, “Generative” refers to its ability to generate new text based on the input it receives, “Pre-trained” refers to the fact that it was trained on a large corpus of text data before being deployed, and “Transformer” is a type of neural network architecture that it uses for processing and generating text.
The answer came! I asked then if it could give a simple explanation of ChatGPT.
ChatGPT features and capabilities
There is no doubt ChatGPT has amazing features and capabilities:
Technical Help
It can answer very quickly complex as well as simple queries. It has the ability to provide helpful suggestions for programmers and code writers. You can for example ask ChatGPT to suggest the best programming language for a specific app and it can also provide quick solutions to your coding problems. Some coders think they may become redundant. It can help you sort out your Excel formulas and data base queries. Try it out and you will see how it can help you. Just search OpenAi in Google. You will arrive at the following site.
In addition to finding out about GPT4 you can try out ChatGPT. It is free and it is the best way for you to see what it can offer as a friend. Undoubtedly this ChatAI is a friend in this context.
Language Learning
As well as providing insightful technical support, another of ChatGPT’s friendly benefits is in Language Learning. As it happens I have a specific interest in learning foreign languages. The language apps DUOLINGO and BABBEL are on my Ipad and I have tried support programs provided by Foreign Universities. I am currently studying French and German on Duolingo and Swedish on Babbel. Trying Spanish on Duolingo was interesting but Latin on Duolingo was very poor. I studied Latin at school for six years so I could gauge its worth. I asked Chat GPT the following question and the answer came very swiftly.
How can ChatGPT help in language learning?
As a language model, ChatGPT can be a valuable resource for language learners in several ways:
- Vocabulary building: ChatGPT can provide synonyms, antonyms, and definitions for words, as well as offer examples of how they are used in context. This can be especially helpful for those studying a new language.
- Conversation practice: ChatGPT can engage in conversation with language learners, allowing them to practice their speaking and listening skills. This can be especially useful for learners who don’t have access to native speakers.
- Writing assistance: ChatGPT can help language learners improve their writing skills by suggesting alternative phrasing, correcting grammar and spelling errors, and offering feedback on sentence structure.
- Cultural context: ChatGPT can provide cultural context and explain idiomatic expressions and common phrases that may be difficult for learners to understand. This can be especially helpful for those learning a language for use in a specific cultural context.
Not a stand alone solution
Overall, ChatGPT can be a valuable tool for language learners seeking to improve their language skills. However, it’s important to note that it should be used in conjunction with other language-learning resources, such as textbooks, language classes, and conversations with native speakers.
ChatGPT can obviously help with language learning and expanding vocabulary. The existing language app DUOLINGO already boasts that it is using AI language technology.
An aid to writing assignments and blogs
It was a great help to me in writing this blog to be able to ask very specific questions about ChatAIs and to have the information I needed instantaneously. ChatGTP in fact gave me an outline of a blog about the pros and cons of ChatGTP. I did not use their exact suggestions but ideas were generated for me. There were also suggestions about suitable images for the blog. One was for a collage of ChatApp logos. In the same way for students with assignments to produce ChatAI’s could write their answers. According to my granddaughter ChatAIs can also help teachers mark them.
However this smacks very much of plagiarism to most people. I asked the question of ChatGPT.
ChatAIs – Friends with Reservations
Technical Help
Above we have listed three benefits of using ChatAIs. In all three examples there are obviously benefits to users. ChatGPT can help with technical issues with writing code and formulae. Some coders are worried they will lose their jobs but others assure them that however good AI is at coding it still needs a human brain to check it out. AI is not an intuitive as the human coder and developer. It is not as adaptable to change as the human experience. But that could change! How do we know AI will not become more intuitive and adaptable.
Language Learning
It is evident that ChatAIs can help with language learning, but ChatGPT tells us that as an aid to language learning we need to use it with other more traditional tools. Google Translate is an excellent support in language learning but it has its limitations with idiomatic speech and can make some very embarrassing errors in translation.
Assistance in Producing Assignments and Blogs.
Obviously ChatAIs can be of great assistance in researching topics. ChatGTP recognises the problem of simply using verbatim what it says.
“It is important to understand that using AI language models like myself is a tool that can used to assist in generating content but it should not be a replacement for your own critical thinking and creativity. When using my output, it is important to use it responsibly and ethically.
Can we rely on everyone to use it responsibly and ethically?
ChatAIs – Are they Foes?
The news about Governments being worried about TikTok and ChatGPT set me off on doing some research on the issue. I did use ChatGPT to tell me why ChatAIs are feared. One of my questions was about TIKTOK and why Governments were so concerned.
ChatGPT ‘s reply to the question about TikTok
“Some governments have asked their employees not to use TikTok due to concerns about potential security risks associated with the app. TikTok is owned by the Chinese company ByteDance, which has raised concerns about the potential for the app to collect sensitive user data and share it with the Chinese government.
In the United States, for example, former President Donald Trump signed an executive order in August 2020 that would have banned TikTok unless it was sold to a U.S. company. The order cited concerns about TikTok’s collection of user data and the potential for the Chinese government to access that data.
Although the ban was later overturned by President Joe Biden, concerns about the security of TikTok and other Chinese-owned apps have not gone away. Some government agencies and private companies continue to discourage or prohibit their employees from using TikTok and other similar apps due to security concerns.“
ChatGPT ‘s reply to the question about ChatGPT and security issues
The ChatAI admits there are concerns about security issues. It says ChatGPT “is not inherently a security concern, there are potential risks associated with its use in certain contexts.”
Disinformation and false news
A big concern is that ChatGPT and other ChatAIs can generate convincing fake text to spread disinformation. You can ask ChatGPT to write in a certain way or in the fashion of a particular person. It can apparently write a passable piece of poetry such as a Shakespeare sonnet. An attacker could use ChatGPT to generate convincing emails that appear to be from a trusted source, tricking people to reveal information they should not or clicking on malicious links.
Influencing audiences with biased information
“Another concern is that ChatGPT could be trained on biased or sensitive data, which could lead to the model reproducing and reinforcing those biases or revealing sensitive information unintentionally.” There is the suspicion that electors could be influenced by news articles and TV programs publishing deliberately biased information.”
Again ChatGPT’s response to concerns is that we should use the app responsibly and “prioritise security and privacy of users.”
Elon Musk’s concerns
How can we trust people to use the apps responsibly? It is because it is so difficult to monitor these apps and the worry that they can be used for nefarious purposes, that Elon Musk and others have requested there be a halt to AI development until effective controls are in place. This may include implementing appropriate safeguards such as user verification, content moderation, and privacy policies, as well as ongoing monitoring and evaluation of the model’s performance and potential risks
Unethical Use of ChatAIs
There have over the past fortnight been examples unethical ChatAIs.
False and misleading news articles
Earlier this week a German magazine published what it said was the first interview with Michael Schumacher former Formula One Champion since his accident. Schumacher received serious head injuries in a ski-ing accident ten years ago and is cared for by his family. This scoop of a story was not as it seemed. It had been produced by a chatbot.
Obviously this was an insensitive and totally unethical ruse to boost magazine circulation. It did not succeed. Retribution followed for one journalist.
Click on the link below to follow the story.
https://www.bbc.com/sport/formula1/65361193
AI apps and the artistic World
AI-generated image wins photo competition
The photo shown below won a Photo competition. The “photographer” refused the award.
What does this mean for the Arts – AI can write poetry and novels, paint pictures, produce photos and images and compose music. No wonder Stephen Hawking warned us about what AI can and will do in the future.
Brave New World
Aldous Huxley wrote “Brave New World” almost eighty years ago. We are getting closer to his world. I think we are right to be concerned about AI. Most of us I think do not have a great deal of faith in man controlling the Internet and AI.
Chat AIs – friends or foes