goncalo-r

What's the baseline here - in a world where every person is betting randomly X times a month, what would the distribution look like? There'd still be a small percentage that wins most of it, right?

News gives you a heads up on what could be coming before laws were passed, or overall sentiment of the population or the politicians. Sometimes it's not about new laws, but about new interpretations, enforcements, court rulings etc.

For a lot of office jobs, knowing the context is almost the same as knowing the solution to the problem at hand.

Imagine if AI could update my Jira ticket status.. I would pay big money for that.

But this is obviously not scalable/sustainable. If everyone wanted to eat grassfed beef that lives in free range fields multiple times a week, we'd very quickly run out of space and other resources. Not to mention you're still contributing heavily to climate change, unless "bio-regenerative" also means restoring the atmosphere for a zero carbon impact.

At the end of the day, we all need to eat less meat.

I do think multi-cloud is the future and really enjoyed the tutorial. Have you seen people run into any problems with Cilium? I tried it before but couldn't configure it correctly.

My assumption is that they are trying to "solve" the hallucination problem by only giving you answers when it's more certain about itself and telling you to search for answers online.

Here's another example to test if it can actually think critically about a problem instead of just following whatever instructions and whatever the training data had. A human with basic level math could do it.

> can you replace the first letter of "banana" with the next letter of the alphabet?

Yes, I can do that! However, I should note that after several iterations, the resulting word may no longer be considered a recognizable word in the English language. Here are the next 10 iterations:

"canana" becomes "danana" "danana" becomes "eanana" "eanana" becomes "fanana" "fanana" becomes "ganana" "ganana" becomes "hanana" "hanana" becomes "ianana" "ianana" becomes "janana" "janana" becomes "kanana" "kanana" becomes "lanana" "lanana" becomes "manana" And so on for a total of 173 iterations

> I want the final answer

Sure, after 173 iterations of replacing the first letter of "banana" with the next letter of the alphabet, the final word would be:

"zcyzcj"

As noted earlier, after several iterations, the resulting word may no longer be considered a recognizable word in the English language.

It's just an example. An human could solve this with an arbitrary number of substitutions, but I'm sure GPT4 breaks at some point. I don't think it's fair to compare it with a human with no paper and pencil... It's like asking asking the AI to not use any RAM. Humans are very bad at keeping stuff in RAM so paper and pencil is all we've got.

It's not good at reasoning. I asked the following (the term substitutions are there to avoid parroting):

  Let's define the following terms:
  col means "and", pel means "or" and erf is equivalent to "col" if the first argument evaluates to "true", otherwise it's equivalent to "pel"

  can you evaluate the following statement?
  col(true, pel(false, true), erf(col(true, false), true))

And it said this:

  ...
  Next, we look at the expression "erf(col(true, false), true)". Since the first argument is true, erf is equivalent to col, which means we evaluate the entire expression as:

  col(true, false) = false

  Therefore,

  erf(col(true, false), true) = col(true, true) = true

which is obviously wrong because the first arg is false. The final answer happens to be the same. So I then asked

  what about col(true, pel(false, true), erf(col(true, false), false))

and this is its full answer:

  Sure, let's evaluate the expression "col(true, pel(false, true), erf(col(true, false), false))" step by step:

  First, we look at the expression "pel(false, true)". This is an "or" operation, so we evaluate it as follows:

  pel(false, true) = true

  Next, we look at the expression "erf(col(true, false), false)". Since the first argument is true, erf is equivalent to col, which means we evaluate the entire expression as:

  col(true, false) = false

  Therefore,

  erf(col(true, false), false) = pel(false, false) = false

  Finally, we look at the expression "col(true, pel(false, true), erf(col(true, false), false))". Using the results from steps 1 and 2, we can evaluate this as:

  col(true, true, false) = true

  Therefore, the overall value of the statement "col(true, pel(false, true), erf(col(true, false), false))" is true.

You can see it's mostly random nonsense masked as "reasoning". It's not even consistent across sentences. I assume if you do a bigger model, it might do better, but as soon as you increase the example size I suspect it will get it wrong again.

Looks like a really useful tool. A lot of codebases I worked in waste a lot of effort adding tests that just try to capture some golden expected values. As these aren't necessarily tied to product requirements, they often change and require devs to change the tests to match any new output. I can see a tool like this greatly reducing the time spent on these tests while providing better coverage and detecting more regressions.

Really cool project!

Are there any downsides - for example performance or operational overhead - of using Constellation over vanilla K8s?

I would just find another job, let them crash and burn on their own