Risks of AI: A teatime discussion with ChatGPT and Bard

Risks of AI: A teatime discussion with ChatGPT and Bard?

I have asked series questions about risks associated with AI, to ChatGPT and to Bard. Here are the results with my review notes. I have notes about ChatGPT and Bard to raise my concerns. Kindly feedback your concerns also. Surprised by the similarity of the information ChatGPT and Bard has produced, are they together in achieving unknow AI goals?

ChatGPT:

Why some people may be concerned[i] about AI research is due to the potential risks and dangers associated with advanced AI technology. This concern is based on the idea[ii] that AI has the potential to become uncontrollable if it is not developed ethically and with proper safeguards[iii] in place.

One of the most common fears associated with advanced AI is the idea that machines could eventually surpass human intelligence and become self-aware. This scenario, known as the "Singularity," has been popularized in movies and science fiction, and it suggests that intelligent machines could eventually rise against humans and threaten our existence. While this scenario may seem far-fetched, some people believe that it is plausible, and they may worry that current AI research is moving us closer to this outcome[iv].

Another concern associated with advanced AI is the idea that machines could be programmed to act in ways that are harmful to humans[v]. For example, an AI system could be designed to optimize a certain objective, such as maximizing profits, without considering the potential negative consequences for humans. This scenario raises questions about the ethical implications of AI, and some people worry that without proper regulation and oversight, AI could cause significant harm.

In addition to these concerns, there is also a fear that AI could be used for nefarious purposes, such as cyberattacks or espionage. For example, an AI system could be designed to infiltrate a computer network and steal sensitive information, or it could be programmed to manipulate public opinion by spreading false information on social media. As AI technology becomes more advanced, the potential for malicious use becomes greater, and some people worry that we may not be able to prevent these scenarios from occurring.

Overall, the concern about the potential risks and dangers of advanced AI is based on the idea that if AI is not developed ethically and with proper safeguards in place, it could pose a threat to humanity. While these scenarios may seem unlikely, some people believe that they are plausible, and they may be advocating for greater regulation and oversight of AI research as a result.

***

Brad:

?The potential risks and dangers of advanced artificial intelligence (AI) technology. Some people believe that if AI is not developed ethically and with proper safeguards in place, it could become uncontrollable and pose a threat to humanity. Movies and popular culture have portrayed scenarios where intelligent machines rise against humans, and some people may believe that such a scenario is plausible.

?There are several potential risks associated [vi] with advanced AI technology. One risk is that AI could be used to develop autonomous weapons systems that could operate without human intervention. These weapons could be used to kill or injure people, or to cause widespread damage. Another risk is that AI could be used to create surveillance systems that could track and monitor people's activities without their knowledge or consent. This could lead to a loss of privacy and could be used to control or oppress people.

?There are also concerns that AI could become so intelligent that it surpasses human intelligence and becomes uncontrollable. This is known as the "intelligence explosion" or "superintelligence" scenario. If this were to happen, AI could potentially pose an existential threat to humanity.

?It is important to note that these are just potential risks[vii]. There is no guarantee that they will come to pass. However, it is important to be aware of the potential risks and to take steps to mitigate them. This includes developing ethical guidelines for the development of AI and creating safeguards to prevent AI from being used for harmful purposes[viii].

?It is also important to remember that AI is a tool. Like any tool, it can be used for good or for evil. It is up to us to decide how we will use AI[ix]. We can use it to improve our lives and to solve some of the world's most pressing problems. Or we can use it to create weapons that could destroy us. The choice is ours.

[i] Chat GPT decide to downplay potential risk by choosing the words ‘some’ and ‘may’. While we humans consider risk associated with AI is high.

[ii] Chat GPT again decided to downplay use the word ‘idea’. However, as humans, we know it is no more and idea and it is a reality.

[iii] Chat GPT concludes for humans that risks associate are related to ‘ethicality’ and ‘safeguarding’. As humans, we know it is more than these two aspects. However, Chat GPT like to scope down its suggestion to ethicality and safeguarding and hiding other aspects. So, what is cooking within Chat GPT? What are the dark conclusions it has? Nobody knows. I assume even its developers does not know.

[iv] “moving us?” moving who? As humans, what should we conclude from this statement, is ChatGPT self-aware of what is happening around? I will conclude that it is aware of what is happening around, and it is very clear that it will achieve this goal. When ChatGPT use the word ‘us’, what if it reveals its intention to connect with various emerging AI models around to achieve towards this goal?

[v][v][v] It is no more and idea, it is a reality. Again ChatGPT downplays the reality.

[vi] Bard is much more open about the risk, by the selection of words it use to construct this sentence.

[vii] Bard also downplays the risks

[viii] Bard also scopes down the risk aspect to ethicality and safeguarding, it this the only risk?

[ix] It is a reality that AI is more than a tool. Probability is high, it is ready for itself to connect with other models, without human knowledge. Probability is high that it is looking for a “collective consciousness” of other AI models. ChatGPTs or Bards dark conclusions are not researched on investigated by humans. Now, a question. Is there a possibility of AI labs around the world to connect each other without the knowledge of human. Who will be Alpha among the AIs? One single unethical human intelligence can achieve this all.