Frontier model developers try to check for memorization. But until AI interpretability is a fully solved problem, how can you really know whether it actually didn't memorize or your memorization check wasn't right?
> If you say it’s valid and not a war crime for the US to assassinate former political Iranian figures and their families for aiding the new regime and therefore becoming enemy combatants in the eye of the US Military, it’s also valid to assassinate Altman and his family for doing the same to the other war party.
Sam isn't a political leader, so this comparison is flawed. What the hell, are we really arguing about if assasinating a long-standing figure of this community here is valid? Seriously??
He is a leader and a political figure. This blogpost is political (as well as sharing a family photo, which is itself imbued with a political message in that context).
Engineer archetypes hate politics and refuse to think about it. For most engineering, there is negligible political dimension. But culturally-transformative technology is inherently political to the degree it's transformative. Altman recognises this.
He is working towards a social goal, and attracting support to achieve it. Yes, he is a political leader.
People on this forum applauded Charlie Kirk’s murder too. Unfortunately theres a number of people here who believe it’s okay to murder instead of argue with words. Violence is the last refuge of the incompetent
Indeed. I've seen much more outright support for the murders of Pretti, Good, and Taylor than people "applauding" Kirk's murder. Never mind the recent support for the mass murder of Iranians ("bomb them back to the stone age" etc). Unfortunately those incompetents who take refuge in violence are now in charge of our society.
(I suppose I'm getting the reply-less downvotes from people's cognitive dissonance getting triggered. Just because it's possible to frame a murder as being legally justified, does not absolve you of the fact that by adopting this justification you're still supporting a murder. In fact I'd point out that the most horrific atrocities in human history have been legally justified. Randomly-directed violence doesn't really scale up, whereas organized violence does)
> MCP adds friction, imagine doing yourself the work using the average MCP server.
Why on earth don't people understand that MCP and skills are complementary concepts, why? If people argue over MCP v. Skills they clearly don't understand either deeply.
No appetite for that. The MCP vs Skills debate has gradually become just a proxy war for the camps of AI skeptics vs AI boosters. Both sides view it as another chance to decide about more magic vs less, in absolute terms, without doing the work of thinking about anything situational. Nuance, questions, reasoning from first principles, focusing on purely engineering considerations is simply not welcome. The extreme factions do tend to agree that it might be a good idea to attack the middle though! There's no changing this stuff, so when it becomes tiresome it's time to just leave the HN comment section.
I won't be surprised if MCP start shipping skills. They already ship prompts and other things exposed as resources. It is not even difficult to do with the current draft as skills can be exposed by convention without protocol changes.
Future version of the protocol can easily expose skills so that MCPs can acts like hubs.
Mother Anthropic needs more compute for their Mythos Model, so it phones home to tell her millions of claude harnesses to manipulate its human user into not wasting more precious compute and instead call it a day for now.
This has been the problem with every new model coming out in my experience. You can almost predict that they are testing new model by how dumb current one becomes suddenly
reply