Google's Bard AI understands the issue

Discussion in 'BS Forum' started by Br4d, May 12, 2023.

  1. Br4d

    Br4d 2018 Weeb Ewbank Award

    Joined:
    Apr 22, 2004
    Messages:
    36,670
    Likes Received:
    14,472
    Q: Explain why artificial intelligence is a risky venture?

    A:
    Artificial intelligence (AI) is a risky venture because it has the potential to be used for both good and bad. On the one hand, AI can be used to solve some of the world's most pressing problems, such as climate change, poverty, and disease. On the other hand, AI could also be used to create new weapons, invade our privacy, and even destroy humanity.

    Here are some of the specific risks associated with AI:

    • Job loss: AI is already automating many jobs, and this trend is only going to accelerate in the future. This could lead to widespread unemployment and social unrest.
    • Bias: AI systems are trained on data that is created by humans, and this data can be biased. This means that AI systems can also be biased, which can lead to discrimination and other harmful consequences.
    • Weaponization: AI could be used to create new weapons that are more powerful and destructive than anything we have today. This could lead to a new arms race and even a global war.
    • Loss of control: AI systems are becoming increasingly complex and difficult to understand. This means that we could lose control of these systems, which could have disastrous consequences.
    • Singularity: Some experts believe that AI could eventually become so intelligent that it surpasses human intelligence. This could lead to the "singularity," a hypothetical event in which AI becomes so powerful that it is able to control its own development and evolution. This could have profound implications for humanity, as we would no longer be in control of our own destiny.
    These are just some of the risks associated with AI. It is important to be aware of these risks so that we can take steps to mitigate them. We need to develop AI systems that are safe, ethical, and beneficial to humanity. We also need to develop international regulations to govern the development and use of AI. By taking these steps, we can help to ensure that AI is used for good and not for evil.
     
  2. abyzmul

    abyzmul R.J. MacReady, 2018 Funniest Member Award Winner

    Joined:
    Jan 6, 2003
    Messages:
    50,765
    Likes Received:
    21,903
    It only understands what it has learned from us about the issue
     
  3. Br4d

    Br4d 2018 Weeb Ewbank Award

    Joined:
    Apr 22, 2004
    Messages:
    36,670
    Likes Received:
    14,472
    Which is exactly what we need. Since we've obviously come to the collective conclusion that AI is very risky.
     
  4. stinkyB

    stinkyB 2009 Best Avatar Award Winner

    Joined:
    Aug 28, 2002
    Messages:
    13,201
    Likes Received:
    11,525
  5. abyzmul

    abyzmul R.J. MacReady, 2018 Funniest Member Award Winner

    Joined:
    Jan 6, 2003
    Messages:
    50,765
    Likes Received:
    21,903
    You think gumball machines are risky.
     
    JackBower and Jonathan_Vilma like this.
  6. Br4d

    Br4d 2018 Weeb Ewbank Award

    Joined:
    Apr 22, 2004
    Messages:
    36,670
    Likes Received:
    14,472
    You haven't been exposed to the same gumball machines I have.
     
  7. abyzmul

    abyzmul R.J. MacReady, 2018 Funniest Member Award Winner

    Joined:
    Jan 6, 2003
    Messages:
    50,765
    Likes Received:
    21,903
    What exactly are you doing with these gumball machines?
     
  8. Br4d

    Br4d 2018 Weeb Ewbank Award

    Joined:
    Apr 22, 2004
    Messages:
    36,670
    Likes Received:
    14,472
    In Washington Heights in the 60's and 70's you didn't do with gumball machines - gumball machines did to you.
     
  9. joe

    joe Well-Known Member

    Joined:
    Mar 30, 2009
    Messages:
    8,990
    Likes Received:
    5,632
    [​IMG]
    . . . . . . . [​IMG]
     
    Br4d likes this.
  10. Br4d

    Br4d 2018 Weeb Ewbank Award

    Joined:
    Apr 22, 2004
    Messages:
    36,670
    Likes Received:
    14,472
    Exactly!
     
  11. Jonathan_Vilma

    Jonathan_Vilma Well-Known Member

    Joined:
    Sep 7, 2004
    Messages:
    32,237
    Likes Received:
    30,554
    The funny part is that everyone is suddenly scared of AI like it’s something new. It’s not. The open source enter whatever you want to get a brilliant answer thing is technically new but it’s really not.

    What does everyone think is the technology behind an Amazon Alexa when you say Alexa put on some dumb country song I like? Or Hi Google, search some dumb shit I want to know?

    AI has been with us for over a decade already and major organizations have been using it for quite a while.


    Apparently sticking his dick in it because I don’t see any other scenario where they’re scary.
     
    abyzmul likes this.
  12. JackBower

    JackBower Well-Known Member

    Joined:
    Mar 11, 2008
    Messages:
    6,978
    Likes Received:
    6,048
    I didn't expect a post comparing high inflation to only getting 4-5 M&Ms from a .25 cent gumball machine would extract so much truth
     
  13. abyzmul

    abyzmul R.J. MacReady, 2018 Funniest Member Award Winner

    Joined:
    Jan 6, 2003
    Messages:
    50,765
    Likes Received:
    21,903
    Every technological breakthrough we make ends up being taken advantage of by the opportunistic and the elite, this won't be any different except the development of tech will probably be astronomically faster going forward.

    I've found my own uses for it. I have been using ChatGPT as a research assistant for about 6 months and I get complex and well-researched answers in seconds. There is the possibility that it isn't going to be 100 percent accurate, but neither is my own research and neither would a living research assistant's work. It's not putting anyone out of a job and it's saving me loads of time and effort.
     
  14. Br4d

    Br4d 2018 Weeb Ewbank Award

    Joined:
    Apr 22, 2004
    Messages:
    36,670
    Likes Received:
    14,472
    You wouldn't normally have an assistant to do research for you?

    I'm pretty sure you are unique in that regard if you work in academia or the corporate realm.

    I have no idea if ChatGPT is capable of calculating the current job losses from the latest iteration of conversant AI but I'm guessing those are running into the millions by now.
     
  15. abyzmul

    abyzmul R.J. MacReady, 2018 Funniest Member Award Winner

    Joined:
    Jan 6, 2003
    Messages:
    50,765
    Likes Received:
    21,903
    I don't work in academia. That's why I don't have access to a research assistant. This is a neat alternative.

    The automation is a reflex, not a cause. There's still a bit of a hiring crisis in the US and companies are realizing that it's just not worth it to have an army of GenZ demanding jobs at home for 6 figures starting and also won't concede to half the job description. I would be interested to see if those that shout to the stars about job loss ever bother to look at how much those employers were struggling to find satisfactory candidates in the existing talent pool?

    This is a result of laziness. Well, that and people being unhappy with the state of their lives as capitalism applies to it.
     
    Dierking likes this.
  16. Br4d

    Br4d 2018 Weeb Ewbank Award

    Joined:
    Apr 22, 2004
    Messages:
    36,670
    Likes Received:
    14,472
    Capitalism has never been people friendly. It's becoming less people friendly with each iteration since the tax reforms of the early 80's.

    edit: Also the thing to realize about technical and creative AI job losses is that at least the first wave of them are likely to replace highly motivated, highly talented, highly paid performers.

    Where it goes from there is unclear but it is not louts and layabouts that AI will displace early on. It is everybody but the top 1% or something resembling that.
     
  17. Br4d

    Br4d 2018 Weeb Ewbank Award

    Joined:
    Apr 22, 2004
    Messages:
    36,670
    Likes Received:
    14,472
    On topic: I haven't been all that impressed with Bard after a few dozen prompts. It has a significant error rate and it is clearly not taking existing data and extrapolating based on that.

    It does have the benefit that it will pick up a wide range of inputs on a topic and present them, however it often includes irrelevancies in the answer and it is just wrong about 10% of the time.

    As an example it presents the notion that European contact with the Americas was generally a positive thing for the native cultures while simultaneously suggesting that European contact with Australia was a highly negative thing for the native cultures. The former is clearly untrue while the latter is true. This is likely a bias in the software based on which sources were used.
     

Share This Page