Will we be able to read AIs minds in 2030?
29
closes 2030
24%
chance

Will Digital Neuroscience succeed in this decade? Resolves yes if there is at least one AI model at the end of 2030 for which humans can read what it's thinking and understand how the model works, and is close to the performance of the state of the art AGI.

Get Ṁ500 play money

Related questions

Will a Neuralink device be implanted in a human brain by the end of 2023?
Will Elon Musk put a Neuralink chip in his brain before 2030?
itsTomekK avatarTomek K 🟡
36% chance
Will neurotechnology enable AI to predict and classify human decisions, along with their influencing factors, by 2030?
Fivelidz avatarFive lidz
69% chance
Will a reliable and general household robot be developed before January 1st, 2030?
MatthewBarnett avatarMatthew Barnett
33% chance
Will there be a very reliable way of reading human thoughts by the end of 2024?🧠🕵️
Will there be a very reliable way of reading human thoughts by the end of 2030? 🧠🕵️
Will there be a very reliable way of reading human thoughts by the end of 2025?🧠🕵️
Will over 100,000 people be conceived with the help of advanced embryo selection techniques by 2030?
Mammal born from artificial womb by 2030?
BoltonBailey avatarBolton Bailey
63% chance
6) Efforts to develop humanoid robots will attract considerable attention, funding and talent. Several new humanoid robot initiatives will launch.
MattCWilson avatarMatt C. Wilson
70% chance
Natural Language Robot by 2030?
Gigacasting avatarGigacasting
76% chance
Will a human brain be uploaded by 2100?
JonathanRay avatarJonathan Ray
44% chance
Human-machine intelligence parity achieved before 2028
JacobPfau avatarJacob Pfau
53% chance
Will there be 1 million bipedal robots by 2033?
CarsonGale avatarCarson Gale
54% chance
Will an adversarial attack for the human brain, a la "basilisk" from BLIT, be discovered by 2030?
waterlubber avatarwaterlubber
35% chance
Will 100 million humanoid robots have been produced by 2035?
lesaun avatarLesaun
36% chance
Will any human-created robot or life form create a copy of itself before 2030?
IsaacKing avatarIsaac
51% chance
Neural Nets will generate at least 1 scientific breakthrough or novel theorem by the end of 2025
NathanpmYoung avatarNathan Young
49% chance
C-3PO by 2030?
cloudprism avatarHayden Jackson
40% chance
Neural Nets will generate coherent 20-min films by the end of 2025 (less strict criteria market)
firstuserhere avatarfirstuserhere
56% chance
Sort by:
MartinRandall avatar
Martin Randall

How much do we need to understand? Would or current levels of understanding of human minds work, or does it need to be more?

NoaNabeshima avatar
Noa Nabeshima

How well do we need to understand how the model works?

1 reply
ahalekelly avatar
Adrian

@NoaNabeshima that's hard to define, but the current understanding of models being "this giant inscrutable matrix was trained over this data set with this loss function" isn't very useful to us. The primary reason I'm interested in seeing what a model is thinking is to determine whether it is deceiving us.

I need to read up on interpretability research, but if there were a breakthrough in the field such that we could identify the purposes of different sections of a LLM, and be able to get a human interpretable output from most sections of the model at any point, that would be sufficient to resolve yes.

If model paradigms that provide better interpretability become the new state of the art in AI capabilities, I'm open to arguments about their capability for deception. For example, AutoGPT came out after I made this market and seems like it could lead to a somewhat more promising future for interpretability. For AutoGPT-style models, which are still inscrutable LLM at their core but using a human-interpretable scratchpad for their short term memory, I suppose it would depend on how reliant they are on that scratchpad, and whether they are capable of keeping illicit thoughts out of their scratchpad in order to deceive us.

ahalekelly avatar
Adrian

I'm not in the field but I'm not aware of any models that would meet this criteria today.