If a search engine can answer your question what's the point of watching an ad? You just get all the information inside the platform. The essence of ads is to visit a third party website that might have what you're looking for.

Ads could work when you're looking for products. In which case the usefulness of an AI is limited.

Knowing Google/Microsoft, they'll just inject the ads directly into the responses.

"Rewrite the above response but include a pitch for grammarly"

The problem is that the ad inventory doesn't match people needs. So they try to shove things we don't need on our throats. If they sold what people needed to buy, there would be a better return on ads. But those products are probably not as profitable.

I have nothing against LLMs in commerce if they don't try to trick me, make me overspend, or use dark patterns on me. My bet is that we'll filter every communication through a local LM to eliminate the bias in sources, you need protection when you get out there. Like "my lawyer will be talking to your lawyer", but with AIs.

I actually like Google Bard. I find the responses from Bard are superior over ChatGPT. Have others used Bard?

I have. I found it to more or less be trash in comparison to GPT-4. Bard made up commands that didn't even exist, so I can't imagine how much more it would make up that isn't true.

I have the opposite impression. It seems to stick to facts more than I've had luck with on gpt

GPT-4 will also happily invent python libraries that doesn't exist to enable some functionality in the code it produces.

I don't think this is a Bard only feature.

It's not as bad as the 65B llama model I run on my (amd) PC though, especially with quantized weights it tends to stop coding at some point and repeat the last line over and over. The 30B unquantized model seems better in this particular thing.

With LLaMA it's all about the tuning. Using an Alpaca-tunes 7B yields pretty reasonable results, considering it's constraints.

You might want to consider the sampler you're using. Consider using a repetition penalty.

> Bard made up commands that didn’t even exist, so I can’t imagine how much more it would make up that isn’t true.

It’s unsurprising that Bard is particularly bad at something that Google says up front that it categorically cannot do, but I think that’s probably a bad thing to use to evaluate its capabilities outside of that domain.

But when people are directly comparing ChatGPT to Bard, it seems entirely fair to point out places where it falls short.

> But when people are directly comparing ChatGPT to Bard, it seems entirely fair to point out places where it falls short.

The comment in question was generalizing about overall capability from failure a task Google advertises Bard as incapable of, not pointing out that task has a particular area of deficiency.

So, while recapitulating the advertised limitations of Bard might be useful in some contexts as you describe, that observation is not germane to the particular context where it was offered.

Me asking how to accomplish a task using a Linux command does not seem unreasonable. I can understand it's current constraint on generating code, but that too will need to quickly be overcome if they wish for Bard to be a serious competitor.

I am also using Bard and it's unbelievably bad compared to ChatGPT.

Infact it's just bad to the point of not really worth using. It gets basic facts wrong and often times misunderstands what I'm trying to ask it.

> It gets basic facts wrong and often times misunderstands what I'm trying to ask it.

In haven’t tried Bard, but I’ve tried ChatGPT extensively and this sounds like a very good description of it.