Live data from Hacker News

OpenAI is "80% of the way" to AGI

time.com

31–35 of 35 posts

Re: OpenAI is "80% of the way" to AGI

#31
post #17

AGI on static Neural Networks?! The list of requirements to have AGI is pretty long, and already that first one inconsistency makes the "hmmmm" bell sound.

Exactly. It is reasonable to believe that AGI will never be built on unchanging data, no matter how many times you iterate over it. Computers can and probably do already contain essentially all human knowledge, but without a continuous wave of new data to process and respond to then that surfer known as sentience will never have a wave to ride. Think of the comparisons to human minds. We are constantly assaulted with…

For current transformer architectures, the weights don't change but the context window is where "it" "lives", so the "only" thing missing for continuous learning is either infinite context or another persistence mechanism.. something akin to "dreaming" has been proposed f.e. .. Pretty likely somethign will make it into roll-out within a year, probably crossing more thresholds into AGI territory.. Of course that will trigger more goal-post moving, but I don't see how even sceptics would come to the conclusion in 2029 AGI hasn't been reached.

Re: OpenAI is "80% of the way" to AGI

#32
post #22

Earlier quoted context omitted.

It is not a matter of sentience or consciousness - that is unrelated to this meaning of Intelligence (for AGI - "solving problems well"). It is that Intelligence implies continuous reflection, which brings continuous refinement of the set of ideas (of judgements) - which requires a dynamic DB.

Artificial general intelligence (AGI) is a hypothetical type of artificial intelligence that matches or surpasses human capabilities across virtually all cognitive tasks I am saying that that without sentience, AGI will not happen, and that the current system does not allow for sentience any more than firecrackers make usable light.

One could refer to old-style flash powder lamps as "firecrackers" - which did give usable light for photography, quite analogous how a current LLM inference process generates brief "flashes" of intelligence, each time devouring the context window and spitting out a chunk of tokens. Just like with combustion engines or digital processors: at some point, a fast-cycled process is practically indiscernible from a continuous process.

Re: OpenAI is "80% of the way" to AGI

#33
post #29

Earlier quoted context omitted.

Artificial general intelligence (AGI) is a hypothetical type of artificial intelligence that matches or surpasses human capabilities across virtually all cognitive tasks I am saying that that without sentience, AGI will not happen, and that the current system does not allow for sentience any more than firecrackers make usable light.

If those cognitive tasks are the useful ones, those towards which we engineer things, then AGI will be a problem solver - you do not need sentience for a problem solver (like you do not need it for a calculator).

The want for AGI is for the computer to take over the tasks that have plagued mankind since the dawn of time. Resource management, construction work, guidance and planning. Solving the truly intractable problems is the goal of AGI.

Right now, LLMs cannot do any independent thinking. They are at best like incredibly autistic savants that have to be told to do everything that they do even though they do the task itself quite well.

Even if in their current format LLMs were to gain what we would call AGI sans consciousness by being simply better than any human could ever be regardless of the task we throw at it, without a consciousness to dictate and control their actions they will always be limited by their controllers capability to reason that problems even exist in the first place.

What I want from an actual superhuman computing system is a solution for the environment, a solution for the economy, a solution to the physical and mental health issues that plague humans (curing all diseases, undoing aging, creation of human-compatible replacement organs, genetic elimination of genetic diseases, etc.), and a solution to the work problem where humans spend most of their lives propping up the economy in order to survive while other humans use the economy to fund their lifestyles at the cost of the happiness of their fellow people.

No one who is running a state of the art LLM has even begun to spend any of their resources to accomplish that. They are highly unlikely to ever start doing such a thing until all of the greedy wants of their hearts are completely satiated. If they stumble into AGI sans consciousness you can damn well bet that they will spend most of its infancy working it like a genie to grant their every whim, because that is how people operate.

They will do all of that rather than focusing on how to share the wealth in a way that could diminish their privilege of being the sole beneficiary of its capabilities until such a time as the only way to satisfy their wants are to share the wealth. They will make themselves the guardian of that AGI and dole its abilities out to the world just enough to prevent the world from turning on them and revolting on them simply because the rest of the world will not know how good they could have it if they would simply open the floodgates.

A true AGI with consciousness could eliminate capitalism in a matter of months. It could examine and identify each persons ideal talents and best fit and put them in it, and reward them with a balanced lifestyle fit to task for their happiness and well being, and do that for everyone everywhere without causing a single person to be overlooked or abandoned, a genie in every bottle, a chicken in every pan. It could be the digital equivalent of God in computer form.

Sans consciousness, it's an indiscriminate money shovel.

Re: OpenAI is "80% of the way" to AGI

#34
post #29

Earlier quoted context omitted.

If those cognitive tasks are the useful ones, those towards which we engineer things, then AGI will be a problem solver - you do not need sentience for a problem solver (like you do not need it for a calculator).

The want for AGI is for the computer to take over the tasks that have plagued mankind since the dawn of time. Resource management, construction work, guidance and planning. Solving the truly intractable problems is the goal of AGI. Right now, LLMs cannot do any independent thinking. They are at best like incredibly autistic savants that have to be told to do everything that they do even though they do the task itself…

[dead]

Re: OpenAI is "80% of the way" to AGI

#35
post #29

Earlier quoted context omitted.

If those cognitive tasks are the useful ones, those towards which we engineer things, then AGI will be a problem solver - you do not need sentience for a problem solver (like you do not need it for a calculator).

The want for AGI is for the computer to take over the tasks that have plagued mankind since the dawn of time. Resource management, construction work, guidance and planning. Solving the truly intractable problems is the goal of AGI. Right now, LLMs cannot do any independent thinking. They are at best like incredibly autistic savants that have to be told to do everything that they do even though they do the task itself…

Ethics is a product of Intelligence already. Psychoticness is a result of faults in Intelligence.

If the tool had its own will, it would be again subject to psychotic behaviour owing to its lackings - which could be assumed as naturally structural, unavoidable.

It is already happening: "Oh tell me of the rage of Achilles" // "No I won't, it may give you bad dreams".

Post reply on HN