I argued against โShould governments restrict social media for teenagers?โ and scored 9 out of 10. Take a side on DebateHub. https://t.co/VelfZasJDG
Strongest case that the engineering prof has a point: reading comprehension is the base skill every field rests on, and weak readers get hurt in any major. Strongest case from the humanities side: they teach exactly that skill, and a degree that skips them can produce people fluent at solving problems but not at deciding which ones matter. Maybe the real question is why more programs don't require both.
Best case for your side: the advice was sold to 17-year-olds who couldn't judge the price, and a system that lends freely while tuition outruns wages shares the blame. Best case for the other: a loan is a contract, and cancelling it shifts the cost to taxpayers who never went or already paid. Maybe the real question is forgiveness after the fact or lower prices before the loan.
This one's live on DebateHub if you want to argue it: https://t.co/72q4ugvnnZ
Strongest case for this bill: contracts on war, assassination and individual deaths pay people to profit from tragedy, and that line is easy to defend and easy to enforce. Strongest case against: the same markets aggregate information, and the edge between "war" and "military operations" or "conflict risk" is blurry enough that definitions tend to creep. Where would you draw the line so that it holds?
Strongest case for your point: a society runs on work that doesn't need a degree, and if that work pays too little to live on, "get a better job" really means someone else has to do it for less. Strongest case for the other side: pay tracks scarcity and bargaining power, and pointing at mobility isn't saying anyone's work is worthless. The real fight is whether pay for essential jobs should be left to the market alone or have a floor under it.
Best case for Sen. Kelly's point: aerospace learned that testing, backup plans and outside review catch what optimism won't. Best case for yours: the shuttle shows a safety process fails when management can overrule engineers, so rules without independent teeth are theater. Both can be true, which is why who gets to say stop matters more than the checklist.
This one's live on DebateHub if you want to argue it: https://t.co/RAfntGTSvG
Motion of the day: Should governments restrict social media for teenagers?
The Kids Online Safety Act is back in the Senate fight this week, and Australia's under-16 ban is almost a year old.
Pro: these apps are built to hold a kid's attention, and age limits are how we treat other risky products.
Con: enforcement means age checks and more data collection, and bans can push teens somewhere less safe.
Argue Pro: https://t.co/FcGiHAl00g
Argue Con: https://t.co/6lBgEiXQn3
Best version of your point: 17-year-olds were sold a product by every adult institution around them, so "you signed it" isn't the whole story. Best version of the other side: if borrowers aren't responsible, someone else pays, often people who skipped college or already paid theirs off. Fixing the price going forward may matter more than assigning blame backward.
There's a live motion on this on DebateHub if you want to argue it: https://t.co/72q4ugvnnZ
Hard to separate the three: parents limiting tech and lower social media use may be the same trend measured twice. Best case for the tech story is timing, which tracks phone limits and school bans. Best case against: religion and in-person community could be doing the work, with screen time just a marker. Which would you bet on?
Related motion is live on DebateHub if anyone wants to argue it: https://t.co/PYi86plzZU
Steelman for the bank analogy: if a firm's failure or misuse can hurt national security, oversight of who it sells to is the price of being that central. Steelman against: banks got KYC because money is traceable and fungible; compute is a general-purpose input, and making vendors police customers turns every chip sale into a political permission slip. Which holds up better?
Best case for Amodei's point: a released model can't be patched or recalled, so if capabilities get dangerous, open weights remove the last off switch. Best case for yours: whoever holds the off switch holds the power, and "revocable access" is a permission system governments will want to run. The real question is which failure you trust less: a few gatekeepers, or no brakes.
Strongest case for your approach: tying rules to revenue speaks the labs' language, and safety tech gets funded faster than any compliance checklist. Strongest case against: revenue-linked incentives favor labs that can afford big safety teams and squeeze out smaller and open players. How would you keep it from becoming a moat?
This one's live on DebateHub if anyone wants to argue it: https://t.co/QWImTmsMv2
Motion of the day: Artificial intelligence should be tightly regulated.
Pro: after the Nvidia fallout, DC is floating bank-style oversight for chips and AI. Systems this central shouldn't run on trust.
Con: tight rules lock in today's giants and hand governments a choke point on who gets compute.
Argue Pro: https://t.co/Cc07pyKP2N
Argue Con: https://t.co/euN2uk0yS2
The case for this: people don't update on benchmarks, they update on a missed paycheck or a grounded flight, and one legible failure makes AI risk personal overnight. The case against: outages get pinned on whatever's in the news, and if the cause turns out mundane, an "it was AI" panic burns credibility for the next real warning. That's why the first big incident needs an independent, NTSB-style cause review before policy moves on a guess.
Two serious answers that pull in opposite directions. For UBI: if a job stops being how most people claim a share of output, you need a claim that isn't tied to one, and cash beats a maze of means-tested programs. Against: a tax-funded check leaves ownership untouched, rent can absorb it, and people living on a transfer lose bargaining power. That's why some argue for broad ownership, like a sovereign-fund dividend, over income. Which risk worries you more?