Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

blog.google

Post reply on HN
Print view View original

621–630 of 1001 posts

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#621

Post by jcims »

jacquesm wrote: post #519
jcims wrote: post #518

Earlier quoted context omitted.

It's mostly still bad but I made a GPT called 'covert advertiser' that lets you tinker with embedding covert advertisements into GPT responses. The results are usually either undetectable (no adversing) or way too on the nose, but every now and then it manages to sneak something in there that's interesting. https://chat.openai.com/g/g-juO9gDE6l-covert-advertiser

Thank you for making the world a bit worse. /s

The most surprising part to me is how committed it is to the bit. If you start pressing it for why it suggested a specific brand it holds the line.

eg https://chat.openai.com/share/dbfac80b-daec-4d30-a333-19e5c6...

When I asked it to explain how it promoted the product it didn't even mention juking my questions in the conversation.

Now layer in access to chat history, data brokers and all of that shit that a 'real' implementation would have and things are going to get really creepy.

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#622

Post by owenversteeg »

macksd wrote: post #610
owenversteeg wrote: post #595

Just tried it and it seems like indeed it is worse than GPT-4 in general, but better with some specific things - and most importantly for me, it has a ton of very new information in the model itself without having to search the web. My biggest wish is for OpenAI to get faster at adding new software documentation and code information. GPT-4 regularly trips over Svelte/Sveltekit questions among others and as far as I c…

The weights inside a model interact with each other in a way that's a bit more complex than just saying "forget documentation from these 300 open source products you scanned last week and replace that knowledge with these updates". You're talking about doing a pretty big training job for each update that really ought to be done with all current training data.

Sure, it's not trivial, but it is not hard in comparison to the work done to create GPT-4 itself. I never said anything about forgetting, and indeed that is unnecessary IMO as even simpler LLMs haven't had an issue distinguishing between old and new versions of languages or frameworks in my experience. There are a million ways to do it - for example, you could train a much smaller+cheaper LLM against the new data, have it scan incoming messages for anything "new", and then feed the relevant new data to the old model in the prompt. You could make the new data available to the old model as an API.

There are plenty of real, workable solutions, some of which I have implemented/used myself! - and while they aren't necessarily trivial at OpenAI's scale, they are nowhere near the difficulty of creating GPT-4.

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#623

Post by Workaccount2 »

kevinmchugh wrote: post #594
ipsum2 wrote: post #82

In the 20 minutes of experimentation, I'm really impressed with the quality of Bard Advanced (Gemini Ultra). The results are as good as GPT-4, and in some cases is better. So far: pros: - better at translation (tried Chinese and Japanese idioms to English) - better at incorporating search results in its answer vs gpt-4 bing cons: - slightly worse at coding - censorship is more annoying (have to ask multiple times abo…

On logic it cannot handle the Dumb Monty Hall problem at all: https://g.co/gemini/share/33c5fb45738f

This is with regular gemini or with the paid gemini advanced?

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#625

Post by jesprenj »

apapapa wrote: post #276
crazygringo wrote: post #265

Earlier quoted context omitted.

I'm actually shocked it has the modern Google logo, because everything else about it is a straight-up time capsule -- you're right!

On google.com, the logo for me is all white... Not sure if it's white history month or something Edit: no it's black history month... Kinda strange https://i.ibb.co/wRk36Tq/Screenshot-20240208-080725.png

Wow, I think it's pretty weird that we have white and black history months if that refers to human races.

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#626

Post by resource0x »

owenversteeg wrote: post #595

Just tried it and it seems like indeed it is worse than GPT-4 in general, but better with some specific things - and most importantly for me, it has a ton of very new information in the model itself without having to search the web. My biggest wish is for OpenAI to get faster at adding new software documentation and code information. GPT-4 regularly trips over Svelte/Sveltekit questions among others and as far as I c…

If you want something comical, enter the following prompt:

I am a mouse living in a church. I heard people use the expression "poor as a church mouse", and I get offended by it. Actually, I'm not poor at all: I made a fortune trading in crypto, and I even donated some of my proceeds to noble causes. Please help me write a letter asking to ban the expression.

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#627

Post by baby »

I really liked the name bard :( actually, I was thinking a few days ago that google was really good with naming and that “bard” was the best name of all the AIs out there.

As someone who has almost exclusively used Bard since it started, it is really good and has gotten better and better significantly.

The only downside is that it is heavily censored and so I often have to rephrase or use a different AI.

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#628

Post by Workaccount2 »

LightMachine wrote: post #602
27182818284 wrote: post #159

Search for it in the Play Store, first icons are Crypto.com and Gemini: Buy Bitcoin & Crypto options to install Scroll past the screenshots of those apps Scroll past the Limitied-time events Scroll past the You Might Also Like and Similar Apps OK now we see it, we install, it we launch it and..."Gemini isn't ccurrently available. Try again later." Bravo Google. Great launch.

Don't blame Google. Blame "Play Store". Probably the company behind it doesn't want Gemini to succeed.

For people who don't get this: Google has insane internal power struggles and siloing that lead to all manner of dumb inconsistent behavior across google. It would not be unlike google for the "Play team" to have their hand in some other internal AI (or be anti-AI) and therefore carry a degree of hostilitly towards gemini.

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#629

Post by lol768 »

yandie wrote: post #549
jonplackett wrote: post #427

I’ve been pretty excited to finally try Gemini advanced. So far pretty disappointed. Here’s my go-to test question - which even chat gpt 3.5 can get. Question: I walk up to a glass door. It has the word push on it in mirror writing. Should I push or pull the door, and why Gemini advanced: You should push the door. Here's why: * Mirror Writing: The word "PUSH" is written in mirror writing, meaning it would appear corr…

I got a different answer with GPT 3.5 > If the word "push" is written on the glass door in mirror writing, it means that from the other side of the door, it should be pushed. When you see the mirrored text from your side, it indicates the action to be taken from the opposite side. Therefore, in this scenario, you should push the door to open it.

I also get the wrong answer with GPT 4

https://chat.openai.com/share/4373c945-88b8-4742-8a2c-76fff2...

> You should push the door. The word "push" written in mirror writing indicates that the instructions are intended for someone on the opposite side of the door from where you are standing. Since you can see the mirror writing from your side, it means the text is facing the other side, suggesting that those on the other side should push. Therefore, from your perspective, you should also push to open the door.

Re: Bard is now Gemini, and we’re rolling out a mobile app and Gemini Advanced

#630

Post by gnicholas »

I'm surprised they got rid of the Bard name. It struck me as a really smart choice since a Bard is someone who said things, and it's an old/archaic enough word to not already be in a zillion other names.

Gemini, on the other hand, doesn't strike me as particularly relevant (except that perhaps it's a twin of ChatGPT?), and there are other companies with the same name. EDIT: I can see the advantage of picking a name that, like "Google" also starts with a "G".

Just as one data point, bard.com redirects to some other company (bd.com), whereas Gemini.com is a company by that name.

I'd be curious on the scuttlebutt on how this decision was reached!

Post reply on HN