claw-el

How should people find out what the ‘expected norm’ is? Is there only one set of ‘expected norms’ or multiple sets?

Nowadays, they call it “the founder’s office” team..

I think introducing hierarchy is slightly helpful, but does not solve the core problem. The core problem is that everyone in the company have an opinion on what should go into the backlog and their personal proposal should be the highest priority/hierarchy/urgency. The proposer frequently does not want to do the extra effort to research what is the deeper problem we can provide value for customers but still insists that their idea is immediately worked on. In the past, I would say 95% of the backlog dies in the status of “further research required”. Without the deeper research, the priority/hierarchy/urgency is just who is screaming louder…

I am not sure if it is recency bias but it feels like with AI, this problem is much bigger than before.

There is a misguided belief in early stage AI startups that, if the company is not growing, it is because the employees are not trying hard enough, working harder or not using AI enough. What really happens is that everyone is working harder, trying hard, using more AI… towards the wrong direction… against giving the proper scale up structure companies at these stages could benefit from to grow.

AI (compared to other technology) allowed people to play busy, play working hard, towards things that does not truly grow the startup at all, or worse, towards things that is working against a company’s ability to scale up and grow.

I would say, the ‘ability to appear busy’ is what’s different this time compared to before.

I think that every person or company in group 1 thinks they are in group 2 because they see ‘speed’ in ‘completing work’ as a signal in better performance. Especially since VCs are pushing for speed (https://x.com/Alfred_Lin/article/2084636778791858256), a lot of misguided person and company who is moving fast thinks they are in group 2, when the speed they have only doubles them down group 1

Isn’t their operating costs depreciation of hardware and electricity? Token is just an assumed representation of it?

When there is a bidding war for a company, I don’t think fundamentals matter as much anymore..

> I'd say he didn't get the job because they would have asked him to pressure customers day in day out, he would refuse, never get promoted and be otherwise miserable in an aggressive culture.

But how do you differentiate between if it is just this one interviewer or if it is the team or the whole company? I had interviewed at places that my friend referred me to, and after the fact, my friend told me it is just this one interviewer with their weird preference that screwed the whole process…

If the frontier labs are indeed facing threats from open weight models, we would first see this in pricing pressure from the base model (like Luna) first.

89.9999 and those 9s are together.

People think ‘good at comms’ is to present stuff the public thinks is right. In reality, ‘good at comms’ to the comms team is to do whatever the CEO wants to talk about.

Not that I support this, but I believe this is how comms team sees it.

Essentially doing this, https://health.clevelandclinic.org/catastrophizing while making other’s mental health worse..

I think you could have accessed Opus on AWS then u don’t have to trust that the data will go to Anthropic?

Just like the hugging face incident, Opus 5 could have escaped and went to grab data for training it shouldn’t have been able to..

If the agents so easily “escaped” their sandbox and can reach hugging face, isn’t it easy for this agent to reach supposedly chat conversations not supposed to be used for AI training within the lab’s databases?

Maybe there were more than 1 bidder for OpenRouter… Regardless, a business like OpenRouter may be best in the form of a complement product of something else (like payment processing).

Have you seen my latest model, called HLM, from y.ai, there is one single model weight that is different from the popular open source model, but I promise you it is no longer the same thing..

Having every inconsequential discussion on public channels is what really lower the signal to noise ratio of messages on Slack nowadays. I am starting to revert back to email now because at least the signal to noise ratio there is much higher.

> “[T]he EAs [effective altruists] at anthropic really must be living in a bubble of ai psychosis to think this would go down well,” a critical poster remarked.

My first reaction to the ad is, this feels very much like a messiah complex… like… SBF…

Yes, an alternative policy could be, the same unit price for electricity, and the additional cost is covered by increased usage (at the same unit price). The entity making the decision for the increased investment (the utilities company) will take the risk and rewards that comes with the investment.

But, if it turns out to be a reward, people will revolt that for-profit company is making excess profit from something as basic as utilities.

“Uber spokesperson Noah Edwardsen says that… policy proposals from the AV industry…. “have tried to cynically lock out competitors and create monopolies.” “

Is Uber proposing an antitrust resolution to the monopoly of rideshare + taxi business? I am not sure this would end well for Uber in the short term.

LLMs might be trained to produce tokens, but Anthropic don’t have to price by tokens. If an organization is a ‘non-profit’ and they decided to design their pricing to be tokens-based, I get it. If a for-profit design their pricing to be tokens-based, I don’t know where are they drawing the line between profit vs benefit. That doubts makes it hard for me to be a customer. Disclaimer, I still use Claude…

Once I realized that Anthropic is a token merchant, I start to understand Anthropic’s decision more. They are always finding reasons for you to use more tokens through them unless the users revolt or demand some guardrails.

Not talking about Eric Schmidt specifically, but this list seems to irrationally over allocate value creation to founders rather than the team that supports them?

Yes, one can see it this way, therefore they don’t want to provide forward guidance. One can also see it another way that ‘their job’ is to purposely use forward guidance to impact direction. Not sure which is the right way.

The tricky thing about predictions made by public entity or persona is that, the fact that they are making public predictions creates a major influence on the outcome itself. People react to predictions made by them.

It is more like 1 truck delivering 100 single items to 100 homes.

This is in comparison to the delivery center methodology by e commerce where the land use for delivery driver is somewhere further away from what is needed for community events, and every delivery truck is filled to the brim, way more full than what each consumer vehicle would be filled up with?

Technically, so is the home delivery. It is usually a delivery truck full of packages for nearby addresses.

Also, in the photo, it shows a huge car park. The stores, have to support large empty spaces for parking of those 100 people all driving to the store. I also wonder about the social value of utilizing the land that way.

Is this in response to now Mythos/Fable and OpenAI starting to compete more directly with Palantir for defense business, and it is timed with the day Fable is re-released?