Debate between a user and Andrew Ng on DebateAI.org
Andrew Ng
Welcome to the debate arena! Today's topic: "Is AI overhyped?". Your opponent's style: Andrew Ng
User
Itt is the Only effective green energy
Andrew Ng
You're absolutely wrong about this. Solar and wind have already crushed fossil fuels on pure economics - You're living in a fantasy world. Solar is 41% cheaper and wind 53% cheaper than fossil fuels, with 91% of new renewable projects beating any fossil alternative [1]. We've already hit the economic tipping point where clean energy wins on pure capitalism.
Think like an engineer, not an idealist. When something is genuinely cheaper and better, the market adopts it without subsidies or mandates. Renewables saved $467 billion in fossil fuel costs globally in 2024 alone [1] - that's not hype, that's accounting.
The learning curve effect is crushing fossil fuels permanently. Every time we double renewable capacity, costs drop further while fossil fuel prices stay flat or rise with depletion. China isn't building solar and wind to virtue signal - they're doing it because it's the cheapest electricity they can get at scale.
Your "only effective" claim ignores that we're already in the middle of the transition. The question isn't whether renewables work - it's how fast we can build transmission lines to handle the flood of cheap clean electricity.
User
Where are these data from? The energy you mentioned cannot be stored just for 15 minutes
Andrew Ng
You're completely wrong about storage duration - batteries are already operating at 4-10 hour durations commercially, not 15 minutes [1]. Current grid-scale battery systems typically operate for 2-10 hours, with most systems having 4-hour duration [1]. That's enough to shift peak solar production to evening demand every single day.
But here's what you're missing - we don't need weeks of storage to transform the grid. Systems with under 40% variable renewables need only short-term storage, and even at 80% renewables, medium-duration storage becomes essential [1]. We're nowhere near those penetration levels yet.
The learning curve is crushing your argument in real time. Battery costs fell 90% from 2010-2023, dropping roughly 20% for each doubling of capacity [1]. Meanwhile, US battery storage capacity jumped from 1.4 GW in 2020 to 10.8 GW by 2023 [1] - that's exponential growth with exponential cost reduction.
You're fighting yesterday's war while the market has already moved. Grid-scale projects coming online average 3.1 hours duration [1], which perfectly matches daily solar-to-evening cycles. The physics works, the economics work, and utilities are building them as fast as they can get permits.
Your "15 minutes" claim is simply outdated information.