Proceed with Caution: Navigating the Risks and Rewards of Chat GPT
We appear to be at the dawn of a new technological age and it's hard not to draw comparisons to the transformative impact of the printing press on society over 500 years ago. Just as the printing press revolutionised the way information was shared and disseminated, artificial intelligence (AI) has the potential to transform the way we work, live, and interact with each other. However, with any technological innovation comes potential risks as well as rewards. In this blog, we'll explore our view of the challenges and opportunities presented by AI and offer insights on how to proceed with caution in navigating this rapidly evolving landscape.
Chat GPT is an intelligent chatbot that helps automate tasks. Its name is the acronym for Chat Generative Pre-trained Transformer. It is, in essence, a language model developed by Open AI which reached a million users in just five days after its launch in November 2022, and has since amassed a large following. Its widespread adoption has received constant media coverage, with figures such as Elon Musk cautioning against unchecked development of AI technology.
At Ethiqs, we are the first in the queue when it comes to the latest technological innovations which might improve our services and operations. However, we also recognise the importance of proceeding with caution and understanding the potential risks and unintended consequences of new technologies. In this blog, we'll share our own experiences with AI technology and discuss some transparency and privacy considerations for businesses and individuals. We'll cover other legal issues such as IP in another blog, as this developing area merits special attention.
Rewards of using Chat GPT:
Chat GPT is an AI tool that can help accelerate your thinking and streamline content creation. We use it like a thesaurus, asking for alternative words and phrases and also to re-phrase a sentence or paragraph or write it in a different tone of voice. While it's by no means perfect, it can be a useful tool for generating ideas quickly and help your creative and workflow juices to keep going.
Chat GPT is highly innovative. As a language model trained on vast amounts of internet content, it can rapidly generate scenarios for consideration. It can even provide alternative perspectives when asked to approach a problem from a different angle. For example, you can ask Chat GPT to answer a question from the perspective of a child, an alien, or a member of a different political party, and it will produce different viewpoints.
Risks of using Chat GPT:
Spread of misinformation
While AI tools like Chat GPT can be efficient and speedy, it's important to avoid relying fully on them for technical or legal advice. Chat GPT can make mistakes and get facts wrong, which leads to the spread of misinformation which can have real-world consequences. For example, in a recent case, a US law professor was falsely accused of sexual harassment by Chat GPT, based on a trip that never happened. It's crucial to use critical thinking when considering Chat GPT's responses, and not to follow its advice blindly.
Data privacy concerns
It's important to be mindful of the type of information you share (whether this is personal or commercially sensitive data) with Chat GPT, as it may not always be capable of keeping secrets. Once you have let the balloon go – you cannot get it back. In a recent incident, a software engineer at Samsung uploaded proprietary code into Chat GPT to try and fix a bug but was unable to delete it once uploaded. This means that Samsung's own code is now part of Chat GPT's learning model and could potentially be accessed by others with the right prompts.
If you are using Chat GPT technology in your website chatbot, there are several privacy considerations. The ICO has recently published a blog which outlines what you need to be mindful of. One of these issues is ‘transparency’ – do you customers know how you are processing their personal data and how you are keeping it secure?
Regulation of Chat GPT:
The various privacy concerns that we have outlined has led to a rise in caution among regulators and governments. Recently the Italian data protection regulator banned Chat GPT and is investigating Open AI ‘with immediate effect’ and their counter parts in Germany, France and Ireland are following this development closely. It may be early days, but the growing popularity of solutions such as Chat GPT means that regulation maybe on the cards. Perhaps even, for the sake of transparency, a way of knowing when these solutions have been used to create content, would be beneficial.
We 'reached out' to Chat GPT (yes really) and asked it if regulation is likely. It said *
“As an AI language model, I cannot predict the future, but it is reasonable to assume that as AI technologies like Chat GPT become more prevalent, there may be increased calls for regulation. Governments and regulatory bodies may feel the need to ensure that these technologies are used responsibly and ethically, protecting user privacy, and addressing potential concerns. The extent and nature of the regulations may vary depending on the specific concerns, the public’s reaction, and how AI developers respond to these challenges”.
Then we asked it to write a Shakespearian sonnet based on recent regulatory updates, and here’s what it delivered:
As AI's embrace becomes ever grander, Chat GPT basks in fame, but doubts persist, Regulations, to protect and ponder, Privacy concerns, cannot be dismissed.
In Italy's realm, Chat GPT must face A ban imposed by those who would protect, In France, Germany, Ireland, a chase, Watchful eyes ever vigilant, inspect.
But as the wave of AI swells and grows, Regulations may shape its mighty force, A world where truth and freedom both compose, And transparency reigns without remorse.
Let us heed the call for openness and light, To guide AI's touch and keep us in sight.
AI-powered technologies like Chat GPT offer many potential benefits and is a reality of our future. However, it's important to approach these technologies with caution and consideration, recognising that they are not infallible and may pose risks to privacy, security, and accuracy. As with any innovation, it's important to stay informed and proactive, and to continue asking questions and engaging in thoughtful discussions about the best ways to use these technologies for the greater good.
If you have any privacy or legal concerns about the use of AI in your business, get in touch with us at firstname.lastname@example.org or book a call here.
*Some of this blog was written with the help of Chat GPT