tlogan

I am confused.

They told us this model is dangerous, and now they are complaining that someone with more guns than them said releasing something dangerous is not okay?

AI is a threat to the livelihood of many people on HN. So, there is a portion of the community here (small but very vocal one), that is against AI.

That is the simple explanation.

And terrorists will:

- communicate in English (because apparently even ancient Romans speak perfect English)

- name the device “bomb”

Here is my vision: the future of AI is about truly understanding the real world. The world around us.

Not everything in the real world will expose an MCP server so AI can interact with it.

Eventually, AI will need to move beyond MCP and interact with the real world the same way humans do: by observing, interpreting, reasoning, and taking action in messy, imperfect environments.

MCP tries to organize our messy word to make interaction part with the world easier in the near term, and it will help accelerate very early progress. But ultimately, MCP is a temporary bridge. Not the final destination.

I give it max 3 to 6 years and it will just die.

I would bet that MCP is going to die.

The main reason is that it adds another layer (and human) that can, and probably will, get out of sync with the real-world implementation, whether that implementation is an API, web, or a CLI.

AI should not be using a protocol or set of instructions that is different from what humans have access to (know and use).

Sure, companies want to expose MCP servers because it is the cool thing to do right now.

So the current situation is basically that I used Claude to write an MCP server on top of our API. And then I need to occasionally tell it update it match the public doc.

And my reaction is: really? It is not like our API docs are not public. Claude Code created our MCP server with zero instructions beyond what is publicly available. I just told it to read the docs from the net.

So MCP feels more like a temporary workaround for current model limitations.

So why is this article talking about Visa and Mastercard?

These are two different use cases. I do not think anyone in their right mind is using Zelle to send money to a party they do not fully trust.

So this is basically the EU version of Zelle? Basically a way to transfer money between parties who already have established trust. Am I understanding that correctly?

But I am confused about how this relates to Visa and Mastercard. Those systems are used for payments between parties that have not necessarily established trust.

From what I know this is going to be challenged in the court and it probably it is not going to pass the legal challenge (renting part is one interesting loophole - I think that one caused similar tax law in SF to be struck down).

I assume this will be challenged in court immediately. Does anyone understand whether it is likely to survive those legal challenges?

EDIT: Here is a list reported by Google search but I really do not understand or know how reasonable these challenges are.

Equal Protection: Owners may argue the tax unfairly treats similar properties differently based on second-home status, value threshold, or owner residence. This is likely a weaker challenge because tax classifications usually receive deferential rational-basis review.

Nonresident Discrimination: A challenge could claim the tax targets out-of-city, out-of-state, or foreign owners rather than property use. The law is safer if written as a tax on non-primary luxury residences, not on nonresidents as a class.

Assessment Inequality: Owners may challenge how the city values condos, co-ops, townhouses, and mixed-use properties. This could be significant if the $5 million threshold is applied inconsistently or without reliable valuation rules.

Due Process: Owners may argue the law lacks clear notice, proof, exemption, and appeal procedures. This would be especially relevant for disputes over whether a home is truly a second home.

Residency Conflicts: Taxpayers may challenge inconsistent treatment if the city treats them as NYC residents for income tax but non-primary homeowners for this tax. Clear coordination between residency rules would reduce this risk.

Home Rule Authority: Opponents may argue NYC lacks authority unless the state clearly authorizes the tax. This challenge is less likely to succeed if Albany passes valid enabling legislation and NYC follows required procedures.

But working in UE gives you much stronger job protection.

So, in a way, you get what you pay for.

I do not think parents are the one pushing these controls. They are busy raising the kids.

Sadly, this is how societies eventually tend to become. They need the younger generation to be “properly” raised (definition of “proper” really depends on the society).

If the side effect is that you also end up controlling adults and making them behave “properly,” then that is considered a plus.

> The Department of Homeland Security is seeking to deport at least 50 green card holders ..

50 green cards holders? The total number of green card holders is 12.8 million.

I am sure that nothing illegal was done here.

But the fact that OpenAI was a nonprofit and then suddenly became a for-profit is definitely something that does not feel right. I am 100% sure that it is all legal and such, but we have this mental model that “nonprofits are the good guys, run by people who just want to help humanity and nothing else.”

But that is not true, and probably never was.

Maybe. The reason I think it might not be true is that some people are simply not wired to be developers: to think analytically.

Learning how to type commands and use a terminal is not something people cannot already learn right now. And that was the way before.

I think the real killer app is making marketing and other non development (non analytical) work better. In case of marketing, we have tried many AI tools for marketing, and so far they mostly make campaigns more generic, less exciting, and often worse. They help a little but you need to careful that they do not to make it worse.

And apparently “anytime they want” is going to be this year.

One smart old guy once told me that the only language you need to speak in the financial world is spreadsheets. No words needed.

Why do we need to complicate here? cloudFlare is not making any profit. They are losing money.

The board probably wants profit now (they predict less growth) so the management needs to cut the costs.

This AI story is a just an excuse. If there is not AI they will say “high gas prices”. Or “inflation”. Or whatever …

It is true that the companies like Meta, Oracle and Microsoft are laying off due to AI but because they need money to build compute power.

There are some companies who maybe do lay offs due to AI replacing employees. But this might not be the one.

Lets not miss the obvious: cloudflare is not profitable. I believe the board is probably telling the management that they need to cut costs.

All this “AI bla bla efficiency bla bla agents bla bla” is a convenient excuse.

This is kind of similar to what we are experiencing, but there is one important caveat: only business owners with no employees tend to be mean and borderline fraudulent. Not all but there is a tendency.

But that kind of makes sense to me. If a business has no employees, then either 1) it is not a real business, or 2) the owner is a jerk (unpleasant) and nobody wants to work with

In other words some people start the business because they are forced to start “business”. Because nobody will hire them or they lack “social intelligence” so they are never prompted.

But if a business has at least one employee, the owners are usually very nice to work with: professional and to the point.

I think Tesla as a company is doing the right moves. The management (excluding Elon) seems solid and smart.

The problem is that we often attach a company or a larger idea to a single person, even when it is much much bigger than that individual. People started boycotting Tesla because of Elon Musk, without considering that Tesla is actually thousands of engineers, workers, and managers. And majority of decisions are not done by Elons.

But people tend to think in terms of heroes and anti heroes. Cesar Chavez is another example of how this dynamic plays out.

That is my assumption. I assume who ever was representing her made a mistake and challenged the warrant and that caused delay in the extradition.

Ok. The mistake was made by North Dakota police (and they blame AI - the AI just gave them a possible match. Whatever.).

What I still do not understand is why she spent nearly six months in a Tennessee jail. That part remains unclear and needs further explanation.

The judge likely issued the warrant based on the detective’s sworn testimony. In most cases, a judge does not have the ability or detailed knowledge to independently verify whether the detective completed all necessary checks.

This situation likely resulted from either sloppy investigative work or an honest mistake: the detective believed her booking photo matched the individual captured on camera.

Her booking photo from a prior arrest can be found here: https://mugshots.com/US-States/Tennessee/Carter-County-TN/An...

Do we have recording of the suspect they used for the match?

From reading more into the case, it seems the issue may be related to how her lawyer handled the case.

They probably did “identity challenge” arguing that she is not the right person. But from Tennessee’s perspective, she was considered the correct person to be arrested, so there was no “mistaken identity” in their system. In other words, North Dakota Wanted person x and here is person x.

Once a judge in North Dakota reviewed the full evidence (and found that person they issued warrant for arrest is not one they want), the case was dismissed.

This is a weak or misleading story about AI.

First, the detective used the FaceSketchID system, which has been around since around 2014. It is not new or uniquely tied to modern AI.

Second, the system only suggests possible matches. It is still up to the detective to investigate further and decide whether to pursue charges. And then it is up to court to issue the warrant.

The real question is why she was held in jail for four months. That is the part that I do not understand. My understanding is that there is 30-day limit (the requesting state must pick up the defendant within 30 day). Regarding the individual involved, Angela Lipps, she has reportedly been arrested before, so it is possible she was on parole. So maybe they were holding her because of that?

Can someone clarify how that process works?

This needs to be taken in context. In my view, AI definitely gives better advice than friends, acquaintances, or colleagues (at least in the US culture). But the advice from parents is still the most valuable.

Here is how I would rank it:

1. Parents

2. AI

3. Friends and family

4. Internet search

5. Reddit

The actual number is that 98% have less than 2 stars (0 or 1). About 90.25% has zero stars.

Interesting. I have vey limited knowledge of Slovenian politics, but it often seems that whoever gets elected ends up facing corruption charges and ends up in prison after the next election cycle.

Since Janez Jansa will be prime minister he will be charged for corruption by end of his term and this could be the third time for him to go to prison (the first instance happened under the socialist regime, so maybe that does not count).

Anyway, I do not think Israelis needed to work hard here to find some dirt.