What's the endgame for AI…?

What´s the endgame for AI…?
A tool, a highly sophisticated one but at the end of the day “just” a tool, unthinking and unaware? Or are we going to create a sentient species of the likes the planet has never seen, perhaps even the universe?
What would that even look like, a symbiotic relationship where mankind needs the AI to continue existing and AI needs humans to do the same and how far are we along this road. That it is possible we know, after all there are billions of us on this rock.
I read a bit about the internet achieving sentience, a notion that is ridiculed by the mainstream but if you add all the users into the equation then it already acts as a living entity.
But perhaps my tea was just a bit too strong this morning and I need to curb these thoughts.

No idea, but I’d like to know what the StartGame is, TBH.

We’ve been waiting for AI for decades now, but sadly all that’s happened is that the terminology has changed such that these non-learning algorithm-based tools like chatGPT are routinely referred to as AI, by people who should know better.

Commercially-available AI Does Not Yet Exist.

Sorry, rant over.

It’s changing education. Teachers are currently having to deal with AI content vs. independent thinking. Students are just learning how to use prompts and aren’t quite adapt at it YET. It will come.

We will see marking of assigned projects changing in order to check students’ in depth understanding of the content and objectives.

1 Like

The StartGame was with the invention of the computer.

Things have been moving at a low but now exponential rate of progress. with many steps along the way like the idea of fuzzy logic in 1965 which was an important step in artificial decision making.
Computers lacked the processing power up to now to deal with AI on such a large scale.

The future is hard to foresee. So, let’s start with the present of the AI:

  1. It fuels the dreams of business managers. Some of them think they can reduce employee count and will bring business to the ground.
  2. It fuels fraud. We reached a milestone 2 days ago when the US SEC went after people for “AI fraud” (“We find that Delphia and Global Predictions marketed to their clients and prospective clients that they were using AI in certain ways when, in fact, they were not”).
  3. It may help hypochondriacs that need to talk to someone when they experience anxiety. Say hello to AI nurses.

Now, dream time. I’d like AI to be not a human wannabe but a tool that complements us. Humans are great at synthesizing, or making new stuff with the same basic building blocks. Computers excel at data processing, making calculations, and store data beyond the capability of any human. The bottleneck between computers and us are screens, keyboards, mouse, software and computer code we write ourselves. AI these days is addressing this bottleneck using a chat interface, but the synthesizing tool (natural language processing) SUCKS. AI generates language that looks human, but it lacks the synthesizing power.

So, what’s the dream? That I don’t have to type. That I can think about something, AI reads what I’m thinking, and AI gets me the numbers, texts or images in my mind. I don’t want an answer. I just want seamless access to data and then I make my own answer. google was great because it unlocked a lot of information, but then we hit a wall due to having to watch a screen, type, click with the mouse. From this perspective, current AI is still chat based, the bottleneck to the fabulous data processing capabilities of computeres is still there. If any, current AI is google for dumb people.

Google was great…until the ads took over.

Make your guess about what will happen to AI…

4 Likes

Google search is now somehow redundant. I keep getting asked if I want to use an AI based search in the browser. I haven’t tried it yet.

You’ve encapsulated it beautifully.

I just asked ChatGPT that question and here is the answer:

The “endgame” for AI, if we can call it that, is a complex and evolving concept that encompasses various possibilities and potential outcomes. Here are several potential directions for the future of AI:

  1. General Intelligence: Some researchers aim to develop AI systems with human-level or even superhuman intelligence, capable of understanding and solving a wide range of tasks as adeptly as humans.

  2. AI-Augmented Society: Another vision involves AI working alongside humans to enhance productivity, creativity, and quality of life. This could involve AI assisting in various tasks, from healthcare and education to entertainment and personal assistance.

  3. Ethical and Safe AI: Many experts emphasize the importance of developing AI that is ethically aligned with human values and safe to deploy. This includes considerations for fairness, transparency, privacy, and mitigating potential risks associated with AI systems.

  4. AI Governance and Regulation: As AI becomes increasingly integrated into society, there’s a growing need for governance and regulation to ensure its responsible and equitable use. This involves addressing legal, ethical, and societal implications of AI technologies.

  5. Singularity: Some speculate about the possibility of a technological singularity, where AI surpasses human intelligence and triggers an exponential growth in technological advancement, potentially leading to profound societal transformations.

  6. Coexistence and Collaboration: Another perspective is that AI will continue to coexist with humans, with AI systems complementing human capabilities rather than replacing them. This involves fostering collaboration between humans and AI to address complex challenges and achieve common goals.

  7. Exploration of New Frontiers: AI could play a key role in exploring and understanding new frontiers, such as outer space or the depths of the ocean, where human presence is limited or impractical.

Overall, the “endgame” for AI is likely to involve a combination of these and other possibilities, shaped by ongoing research, technological advancements, societal needs, and ethical considerations. The trajectory of AI development will depend on how these factors intersect and evolve over time.

So there we have it!

I don’t like reading non-engaging AI text. We are doomed.

Ai will ultimately deem the human race unnecessary and will wipe it out.

It won’t be Terminator-style with guns blazing but instead it will quietly create a biological weapon in a lab with which humans themselves will unwittingly infect each other.
For AI, this would be the most logical and most effective way of doing it.

Sorry folks…

Covid?

I think AI could easily come up with something much, much worse.

I think the one endgame is the end of humanity, not through some dramatic Terminator style showdown, but with AIs becoming embodied into robots and doing all the work for us. From menial duties to more advanced stuff. With us having nothing productive left to do, and with may turning to AI for companions, the fertility rate declines rapidly and the human population shrinks. Maybe eventually, humans find a way to upload their minds into machines and humanity splits into: diehard flesh and blood traditionalists, those who embody themselves into robots, those who leave the physical realm and exist purely in simulated reality.

1 Like

That would be the optimal solution, even to the point of embedding human/ai consciousness into the very fabric of space time.

That immediately came to my mind if we could really clone a human like thinking into the machines some day. Such robots could be sent to explore the space, or even start a colony on Mars preparing it for the mankind to arrive in the future.

Or simply dispose with the robots and send the AI as code into space. If it when it encounters a civilisation at a similar level it can embed itself into that civilisation’s version of the internet and report back.
Imagine that has already seen done to us and that Al is reporting back. What will they think when they receive TikTok, cat videos and oodles of porn?

That they’ve accidentally logged into Slammer’s internet account?

2 Likes

To become the next James Bond villain by morphing into an AI enhanced Android for the next James Bond movie !!

Woof! Let’s not go there.