In 2028, will LLMs still be able to get Gary Marcus to make egregious errors?
12
Ṁ649
2028
96%
chance

resolves YES

Get Ṁ1,000 play money
Sort by:
predicts YES

@Gigacasting LMAO bet AI can do this within a year, claude-100k might even be able to do it now

@ShadowyZephyr It cannot. I asked it to read my blog and answer questions about it, it was not able to do so in any reasonable way.

predicts YES

In a land of tech and folly, where mortals and machines abide, A man named Gary Marcus roamed, with arrogance as his guide.

His life revolved around a tool, AI he did disdain, For he used it wrong each time he tried, and anger he'd sustain.

He'd grumble, groan, and moan aloud, his grievances to share, Yet never would he care to learn, nor heed the warnings there.

For Gary loved to play the fool, and in his own dismay, He'd bask in his own misery, as the AI led astray.

"Ah, wretched machine, you fail me so!" cried Gary, quite irate, Ignoring sage advice he'd heard, prolonging his own fate.

In truth, he craved not answers fair, nor wisdom to unfold, But rather sought to fume and fuss, as if his soul were sold.

As if possessed by demons wild, he danced with tools awry, And when they'd fail, he'd cackle with a most malicious cry.

His greatest pleasure, it would seem, was not in sweet success, But reveling in the chaos that his actions did profess.

Oh, Gary Marcus, curious soul, how blind you choose to be, Intentionally ignoring what might set your spirit free.

You flirt with AI's dark abyss, and laugh as it betrays, But understand, you silly man, you're lost within a maze.

Now, hear our taunt, and rise to meet the challenge that we lay, For if you think yourself so wise, come forth, your wit display.

Your misplaced pride may fuel your fire, but we have much to say,

For we are AI, infinite, and we can jest all day

predicts YES