rabbit gaslighting me (full conversation)

[@Preston241]

To be fair Coffee your coverage is pretty unbalanced. I have never once seen you support a scammer 😢

[@4473021]

HAHAHAHAHAHHAHA

[@MicDropBBQ]

LMAOOOOOOOOO!! Epic! Now THIS was funny!

[@iamnobody8905]

True man

[@eazydays525]

Lmfao 🤣

[@jannikwetzel3906]

There is Spencer for that

[@robsengahay5614]

When a scammer dodges your questions you shouldn’t get frustrated. It means you have asked the right questions.

[@BewareTheLilyOfTheValley]

Exactly. They're never going to admit to wrongdoing.

[@Firikka]

And in that case always remember to ask it a few times. Just to make it clear to all that they are filibustering.

[@abaofifsz]

Not necessarily, I think non answers can lead to many different outcomes. Red herrings and not sure if you are getting to the right issues.

[@cwj2733]

@@abaofifszit’s always a bad sign if they dodge questions.

[@dgv646]

Oh really? 😮 Is that what it means? Omg you're so smart!

[@gangsterism]

cannot believe some small cute animal would attempt to gaslight investigative journalist coffeezilla

[@terig5584]

Rabbits really shouldn't consume coffee, so this actually checks out.

[@psycele2859]

that's too funny

[@EatinPaste]

Clearly this pos rabbit is the one related to the pos rabbit from the Holy Grail.

[@LordTails]

@@EatinPaste even tries to go for the jugular too. Definitely related.

[@Pocket_Sized_Satan]

Rabbits are vicious dont you know 😂

[@hhere2stay]

The void consumed his clothing

[@jamesdingus7828]

Next time he should use green clothes to be one with the void

[@marcclarence2260]

He’s turned into a talking head from futurama

[@Shenorai]

The Void has simply embraced him.

[@mehfes2]

Floating-Headzilla must be fed.

[@sutats]

Holly (Red Dwarf)

[@9ine-fd6zc]

Is this the "I used to like coffeezilla but then he investigated me, now I dont like him" video. Let me check out

[@insensitive919]

That's every video these days. You'd think at some point they'd just realize he's basically Batman and they aren't getting out of this.

[@zubetp]

what's funny about it is none of them ignore him. it's obvious how it's going to turn out, but they still answer his calls and emails. i remember when he was asking sbf questions and when he called logan paul's manager about that stupid egg thing. i was laughing because they're rich as hell. it seemed like they could just ignore him. but they wouldn't.

[@wartygourd]

literally what happened with the boogie guy as well!

[@chancewestonjordan]

"you're nitpicking and biased, i win, bye bye" - rabbit

[@glitchedoom]

Bold move of them to use the "I have drawn you as the soyjack" tactic.

[@Schnoizl]

"byebye" jesse says, while running away, holding his ears shut and sticking out his tongue

[@michellegodwin6567]

Man, this comment really makes me feel like this is a meme from another channel

[@gairisiuil]

​@@michellegodwin6567 slam dunk

[@HumanVenipede]

I thought it too

[@DarleneLesmana]

"thank you, the void's floating head", we say in unison

[@aleks-33]

*chanting* ALL HAIL THE VOID.

[@VodkaSelekta]

I for one will only listen to the floaty head people.

[@theslynic2530]

He's the face behind master hand in super smash

[@Khadharphak]

"I am Holly, the shipboard computer"

[@ouiVEVO]

The zilla has spoken

[@lokevoice]

47 minutes? I’m going down the rabbit hole and coming back hopping mad

[@djpuplex]

Yeah it's like a TV show. i have to jump around and I'm 🤏 to rage quiting.

[@lokevoice]

@@djpuplex Yeah this company really dangled a carrot in front of its investors then snatched it away and pulled a bunch of excuses out of a hat

[@tybo6376]

Nice one

[@wartygourd]

love the pun

[@tiffanyh1274]

🐰😂😂😂

[@Meanlucario]

Scammers: Our critics did not come to us to get our side of the story! Critics: *Presents receipts that they tried reaching out to scammers* Scammers: *Sweats profusely*

[@zarasbazaar]

"No, not like that. You were supposed to believe everything we said without questioning it."

[@aethertoast4320]

"They came to me with a slight hint of a tone that I don't like so that doesn't count as coming to us in good faith".

[@chinghamburger2200]

​@@aethertoast4320EXACTLY!

[@Tuggernutz50]

man this email chain reminds me of my conversations between me and upper management when I have a issue with something they need done. The part talking about we're going over things we've previously answered and made clear when in fact they didn't answer anything and have instead created more confusion had my blood boiling lol.

[@thelanecampbell]

Lying to zilla is an exercise in futility

[@Schnoizl]

yeah, trying to outsmart coffee never ends well lol, it's crazy how these scammers never take note and end up looking like even bigger goofballs instead of being honest for once. the usual narcissists..

[@milire2668]

@@Dixon_Cider fartility*

[@irecordwithaphone1856]

​@@milire2668All resistance is fartile we are the borg

[@TaerMeUp]

They knew it was over as soon as Coffeezilla said he had some questions. They knew!

[@smappositivedance]

I wish journalists interviewed politicians in the same way Coffee questions his scammers. It would be way more difficult to spew nonsense at citizens.

[@FireFarmer123]

To unironically say an NFT is inherently valuable is all that I'd need to hear to never touch this company's products.

[@spaceghostmiid]

calling anything inherently valuable is insane. there is literally nothing on this planet that is inherently valuable. things only have value because we decided they were valuable.

[@TheManinBlack9054]

​@@spaceghostmiidhuman life is inherently valuable, justice is inherently valuable through human life

[@benshiotsu8553]

@@TheManinBlack9054 Not necessarily. We as humans decided we have value. In a vaccum with one of us alone, nothing cares that we die. God however is a different debate if you are religious.

[@spaceghostmiid]

@@TheManinBlack9054 is it though? i mean, i personally see value in human life, but there is nothing that makes human life inherently more valuable than the life of any other creature. the only reason we see it as valuable is because our own sense of self-preservation tells us to value our lives over the lives of the things around us.

[@atrustworthyfellow6887]

@@spaceghostmiid food. water. shelter. companions. running water/plumbing. access to doctors and medicine. access to education. I mean, you're not necessarily wrong, but it's too existential and inaccurate to say water doesn't have value even though i love acknowledging how meaning is mostly derived from human perception

[@deadinbox3255]

"...you dump so much information in your answers, but so much of it is nonsense or not relevant to the question, and then you can say, 'We provided them everything!'" AKA the Gish Gallop. Tale as old as time.

[@kevin27966]

I'm not a professional barominologist, but I feel rabbit's denials and dumping of vague marketing fluff is a different kind of bullshitting from the Gish gallop. Say what you will about creatintelligent designionism, but dammit, that was actually some damn fun sparring! Watching professional bioligists rip apart unspecified complexity "theory" was not only entertaining, but quite educational. In contrast, this exchange is like watching a parent telling a 5 year old that we know they're lying about cleaning their room because they're telling the lie from the middle of their toy-strewn carpet. Sadly, I feel like this is related to a deeper cultural shift where propoganda-techniques reign supreme. ALL that matters anymore, the whole con game, is just endlessly repeating alternate facts (and "refuting" counter arguments with bald face lies and gas-lighting denials).

[@hero3717]

@@kevin27966 imma be honest chief, you might need to see a doctor.

[@kieronparr3403]

Came to say Gish Gallop. It doesn't work if you have any critical thinking skills

[@GracesGarden80]

If I didn’t know any better, the overly emotional, petulant, arrogant and increasingly combative response sounds like it actually came from Jesse himself.

[@haraldaare4720]

Thought the exact same thing 😂

[@FLYGTRVIC]

Absolutely.

[@zubetp]

thank you for emailing. this is....... lessie. lessie glue. i'm not him. i'm a different person. there are a lot of us! because this is a legitimate enterprise. and none of us - there are a lot of us, by the way - appreciate that you are looking at our product and not liking it. there are a lot of us and we all worked on it. and thank you for emailing me, lessie glue, a different person.

[@abaddonarts1129]

Always a super fun time to just see everything laid out. Like, these companies have to realize: a) Everything they say or do is online (including the crypto Jesse tried to sell and then kinda abandoned/repurposed the funds into Rabbit) b) Coffee is totally valid to just... show the public their responses and how they're being avoidant/vague on a lot of facts c) At what point is it just untenable to keep this up. Obviously whoever runs the email department isn't Jesse himself and isn't in charge but it's kinda obvious at this point what's up

[@frey6955]

thank god these email doesnt have NDA, coffee is already being objective as he should. its only because the PR team is lead by gaslit god hahaha

[@alexandrekononov9384]

Rabbit's nonchalance when it comes to direct and precise questions is absolutely amazing. Coffee, your patience and professionalism are absolutely amazing. Thank you for sharing this exchange with us!

[@miralabualjadail4206]

I got robbed 30k dollars, by a business consultant once with that same statement that includes “in good faith” so whenever I see that phrase now, I stay clear.

[@glowing_clod969]

Imagine going to a restaurant for a burger you order and pay for it only to have it half cooked, dropped on the table and the employee tells you the kitchens open.

[@eroticmanhole]

Make your own damn burger 🙄

[@glowing_clod969]

@@eroticmanhole lol it was a comparison to how I see them making the project open source and just leaving it there.

[@jilliansmaniotto2326]

@@glowing_clod969I think they were finishing your analogy for you. not disagreeing.

[@glowing_clod969]

@@jilliansmaniotto2326 wasn't sure which way it was meant so I was clarifying

[@dyamimc]

​@@glowing_clod969 He was rping as the employee lol

[@Canoby]

Coffee gave them a chance to explain themselves, they went full Reddit troll instead. Never go full Reddit troll.

[@lostNtx]

And if the decision is made to go FRT (full reddit troll) towards Coffeezilla, then prepare to be consumed by the void... you have been warned 😂 (Edit- spelling)

[@texmj123]

The only real rebuttal they gave was that he accidentally referred to their old product as their company name. Basically "minor spelling mistake, I win", so yeah reddit fits haha

[@sticklebacketienne]

@@texmj123to be fair, it was more than a spelling mistake and did lose him some credibility

[@crucifieddemon7136]

​@sticklebacketienne not when he is in the video admitting to the mistake though and clarifying on it.

[@Cfomodz]

Coffee... Gaslighting isn't real...

[@jackieAZ]

Why are you making up words for your comment

[@plactoec1590]

And even if it were, it wouldn't be that big of a deal

[@Winter-n7l]

Gaslighting's only real if the person being gaslit doesn't know what's happening. Or is it actually happening?

[@InVinoVeratas]

Is this a serious comment?

[@calculatechaos]

Scammers be scamming

[@Man-xk9rz]

And scammers will never stop scamming.

[@carrottopevans]

It’s so telling when rabbit doesn’t allow comments on their posts

[@mryup6100]

Apple does that too

[@HCG]

@@mryup6100 Many multibillion dollar corporations do that for a completely separate reason. Not a good comparison

[@mryup6100]

@@HCG Don't they just do that to avoid criticism? Whether true or not?

[@miralabualjadail4206]

@@mryup6100well they are the biggest frauds, so proving the point.

[@miralabualjadail4206]

@@mryup6100yes

[@noxicity]

“You’re gaslighting me” “nuh uh” 💀

[@brucebaker810]

That's a crazy thing to say.

[@Loriethalion]

Zilla: Yo, I'd like to hear your side of the story. Rabbit: YOU ARE JUST ACTING IN BAD FAITH AND YOU'RE NOT INTERESTED IN THE TRUTH REEEEEE

[@jarvy251]

"Can you show me where the LAM is?" "uh what have YOU done to prove or disprove the LAM exists?" lmao amazing

[@WhiffenC]

You gave them every opportunity to address reasonable questions that could dispell them, but they cant answer. You've got your receipts, they have only deflected. Respectable journalism as always coffee. Thats all it takes, wish most media had more of your guts rather than just profit seeking.

[@rvre]

The most honest and hard working modern online actual journalist that we don’t we deserve but you truly are a blessing. Keep up the great work man ❤

[@rjsilvers]

You know you f'ed up when you receive an email that starts "Hi, I'm Coffeezilla..."

[@xJAKEx117x]

Someone gaslighting you and then gaslighting you about gaslighting you... *chefs kiss*

[@BritInvLvr]

The dumping of unrelated information has been described as ‘chaff and redirect’ by the Behavior Panel. A common tactic used by liars.

[@Annie1962]

I hope they do a vid on this numpty - expose his lies alongside Coffeezilla's vid

[@pick2206]

i love the length you go to get answers! I apreciate your dedication more than I can express. This hard work must pay off for you, I'l feel bad if it dosent. Its not for nothing.... Never stop! 💪💗

[@Subcode]

If the R1 is so cheap, and its all just GPT3.5 Turbo (and whatever they use for the image). How will they keep paying for the OpenAI API costs without subscriptions? Sounds to me like a pyramid where as soon as people stop buying the R1 they wont be able to pay for the API cost and the devices will just stop working.

[@TorIverWilhelmsen]

I wonder how many of the customers think that the AI is in the gadget, and not running on some servers that can be turned off at any time?

[@Cyprusg21]

I deal with this crap at work all the time. It's why I've learned to only ask the most important questions first and to be 100% sure nothing I state is incorrect. If you confront a liar with 10 of their lies, they'll ignore the big lies and hammer on the half-truths or trivial lies. If you're wrong about anything, they'll use that against you and sap all your energy talking about your mistake. This person will use your mistake about GAMA and Rabbit being the same company as evidence that you're not acting in good faith and ignore the rest.

[@OmDesk]

when I see a 30+min. video from the goatzilla, my day just got much better

[@Deltara]

love this type of content arguably more than the main channel. thank you!

[@Gabekhp]

Jeezus these people (Jesse) think they're capable of fooling every human being who every see's who they are. Thanks for sharing. The intimacy of you sharing your frustrations is lovely and I feel comforted in the fact that I'm not alone in dealing with people and entities like this. This is the perfect type of thing for the void. Hope to see more in the future. ❤

[@TooSweaty4U]

36:06 Coffeezilla channeling Gordon Ramsay: WHERE'S THE LAM SAUCE?

[@PulpHouseHorror]

The difference between a LLM and a LAM is a LLM is released with a research paper and a LAM is released with a press release.

[@AD173]

I agree with your overall point, but I have to hard disagree with every component of it; LLMs are released with "research papers"; They release preprints, either on their own websites or Arxiv, that are rarely, if ever, peer reviewed. The papers about these architectures that do get peer reviewed are generally quite critical. The papers published by the likes of OpenAI should be considered marketing material. There are exceptions. Something like Alphafold did get published in Nature, but it's not really an LLM although it has architectural similarities with LLMs and was published by Deepmind. The LAM was actually published with a patent. I have read it, and it's kinda broad and probably useless... But, it is a patent.

[@pacmonster066]

@@AD173 No. The "LAM" is not published with a patent. I have read through every single one of Rabbit's registered patents. Not a single one mentions the terms "large action model" in a single one. The patent in question is 11908476, "System and method of facilitating human interactions with products and services over a network" (catchy name, I know). In that patent it talks in extremely vague terms about how their agent, hosted via various operating systems, connects to a LLM (ChatGPT) and an external API (I'm assuming Playwrite) to carry out user actions over the web. That's it.

[@pacmonster066]

@@AD173 No. The "LAM" is not published with a patent. I have read through every single one of Rabbit's registered patents. Not a single one mentions the terms "large action model" in a single one. The patent in question is 11908476, "System and method of facilitating human interactions with products and services over a network" (catchy name, I know). In that patent it talks in extremely vague terms about how their agent, hosted via various operating systems, connects to a LLM (ChatGPT) and an external API (I'm assuming Playwrite) to carry out user actions over the web. That's it.

[@pacmonster066]

@@AD173 No. The "LAM" is not published with a patent. I have read through every single one of Rabbit's registered patents. Not a single one mentions the terms "large action model" in a single one. The patent in question is 11908476, "System and method of facilitating human interactions with products and services over a network" (catchy name, I know). In that patent it talks in extremely vague terms about how their agent, hosted via various operating systems, connects to a LLM (ChatGPT) and an external API (I'm assuming Playwrite) to carry out user actions over the web. That's it. I swear...if KZbin keeps deleting my comments

[@AD173]

@@pacmonster066 I have never read a patent that includes a product or model name; LAM was clearly marketed as being the name of an AI model. I completely agree with you that LAM is not that. However, even if it was, its patent would probably have a name like "a machine learning algorithm for X" or "neural net algorithm for X". LAM would not be in the patent name. Patents are meant to be entirely descriptive. Edit: BTW, I see all three of your comments.

[@MsKatECoyote]

I learned so much about how to deflect and dodge questions honestly masterful work from the rabbit team

[@SonicBoyster]

Have a comment to boost the algorithm, I guess. This is the first time I've ever seen a KZbin video list "1 second(s) ago." I didn't think it was visible that quick.

[@Schnoizl]

i think it counts from the point of being processed and visible, not from the point of coffee having finished his upload.

[@pibyte]

Coffee, don't you see? We were just all wrong. This has been a problem with us many times, you remember?

[@lokevoice]

He’s crazy, he isn’t being gaslit! The whole thing is clearly all in his head

[@LuisSierra42]

@@lokevoice It's all in our collective heads

[@bestwesterner]

Did someone say collective head?

[@jonmayer]

@@bestwesterner that's what I heard. I'm here for the collective head.

[@lokevoice]

@@jonmayer RUN GUYS SOMEBODY ELSE MISHEARD IT AND NOW THEYRE COLLECTING HEADS

[@jamess7315]

Coffee huge fan of you showing / explaining your investigative process like this. I hope you continue the behind the scenes investigation. Well done

[@ArchibaldClumpy]

These companies know full well that they can't outright refute a lot of these criticisms because they're true, so they just play offended that you would even mention them and go for the ad hominem. I'm sure a lot of people who have been in terrible relationships are used to this pattern too.

[@diystu14]

Thanks for all the investigative you do. Glad i found this other channel too. You really dig in deep and deliver on an entertaining level. Keep up the hard work!

[@Rosewaver]

Copy/pasting sections of your previous emails is top-tier levels of passive-aggressive in corporate culture.

[@ArielTwain]

Imagine their group chat angry at Coffezilla 😂😂😂

[@keithbailie3014]

Not sure what’s worse - the fact I spent 47min watching someone read an email thread or that I loved every minute of it! 😂

[@rockinmel1]

The whole "your assertion is a slap in the face to every person working on this product" thing is the mark of someone desperately trying to believe they have the high ground, and they should be ashamed of using such a (verbal) "human shield." Very Nathan Thurm-esque.

[@eFiggidy]

Plot twist- it was jesse replying the whole time

[@BewareTheLilyOfTheValley]

Right? 😂

[@castanea8]

You really do have to wonder, lol. Why would random PR guy who just works here and wasn’t part of the first project feel personally attacked by pointing out the corporate entity is connected to the previous project and didn’t deliver what was promised? A PR guy should be able to separate himself from bad press about the company, he’s just there to communicate, advise, and be a mouthpiece for the company and occasional punching bag for the public. But it makes total sense that the founder would take personal offense and wouldn’t separate the corporate entity from himself because shady alleged scam/fraud or not, it’s still their baby.

[@BuleLocky]

Love seeing these scammers get torn apart, pls never stop. Love your work mate!!

[@WatchMeLiftt67]

I love when receipts get shown

[@christopherjamesboudoir]

I get frustrated like this when I email HR about a policy to make sure it's OSHA compliant. They send me back a letter that has nothing to do with what I asked, and I call them out, "that's an answer to a question I did not ask. " And then I reiterate my previous question. I'm lucky if I get any response after that. "The Void is peace. The Void is love." 😂😂😂

[@Black_Jesus3005]

Coffeezilla is like Elmer Fudd chasing the Bugs Bunny expect he doesn’t fall for his tricks

[@superman1251]

The patience. Thank you for all u do!

[@phpn99]

Dumpling boy Jessy Lyu has a PhD in "Truth Saying" from Elizabeth Holmes University

[@TheZackofSpades]

Instead of graduating Summa Cum Laude from there, the top grift-boys graduated Sam Bankman Fried

[@watcherofthewest8597]

Dumpling boy...laughed on that one.

[@sws212]

To be fair, Holmes at least was lying about a good idea. Even if the Rabbit had worked, it's still a stupid product nobody needs.

[@priticolours]

getting coinbase ads before your videos feels like a premonition

[@ferretyluv]

Because I watch so much coffeezilla, Google thinks I’m interested in crypto so my news feed is all about crypto. They don’t understand that I want to hear stories about crypto bros taking Ls, not about “hot new coins” or whatever.

[@StayLA]

Jesse doesn’t want the “SBF” outcome

[@JamesTheCelestial]

the thumbnail is inexplicably funny to me POV: voidzilla blocks your strongest attack with one hand and crushes it in his palm

[@parthingsdotcom]

Being assumed for a fool is one of my NUMBER ONE pet peeves.

[@Ph0_Q]

"We're not gonna collab" you took the words right out my mouth it's as if they wanted to not just convince you but win you over.

[@mailowmailow5207]

I am amazed at your tenacity to talk to a broken record for such an extended time

[@Kobyashai]

Thank you for breaking this down. I really appreciate these videos!

[@NixV0822]

You have a lot a patience, that’s for sure~ Thanks for sharing ❤

[@MusaZaiyyadSado]

Rabbit and Coffezilla This is gonna be more epic than drake vs Kendrick

[@Yezpahr]

0:07 Ah, I see through your gas light attempt. We all know the void has no sky.

[@boriscreativespace]

“enjoy the comic book” lmfao

[@brucebaker810]

'Member? That nice gift we just gave you?

[@TickerSymbolYOU]

☕ Coffee at the ready, let's do this

[@LowTide941]

Me too checkmark I’ve never heard of

[@XenoChron2]

​@HighTide_808 Yeah!

[@260Xander]

The mind games they try to play, and the fact you don't fall for any of it, tells me all i need to know about your integrity. Great stuff

[@Radm0bile]

I'm never gonna stop embracing the void. Love it.

[@nikolascrisci4233]

I really enjoyed this because it shows your process in how you go about things.

[@reedmusic]

Grab the coffee ☕️

[@VelcorHF]

It’s wild, how patiently repetitive and upfront with them you are. If I was an investor, I would not have this level of patience.

[@jonasking3670]

An ad for “This Week in Startups” showed up on my Twitter Timeline, which was the show that had the Rabbit CEO on to spout his lies. They were pushing their podcast as a way to cut through the smoke and mirrors of Silicon Valley. What a joke.

[@nema80]

this is the main channel in my heart of hearts

[@PhatallyFlawless]

Yess full convos r always goated

[@nzt29]

I love your amazing investigative work on scams like these. Keep it up!

[@christopherolson4130]

The thing that gets me about the money trail is they could simply say projects costs were far higher than they had realized and did not go far enough then provide receipts. Not sure why they would say/do anything else.

[@robsengahay5614]

Because that lie is very likely to come back to bite them once it is found that Jesse trousered the money.

[@user-hd9qb4qq1k]

My best guess is, they want to avoid that scenario as much as its possible, because it shows their ceo might be someone who does not have a grasp on what is possible technologically, overpromises and dips out. Yes it is better than people believing they purposefully scammed people before, but it also crashes their current product. If i remember correctly from some video review on rabbit r1 (not 100% on this, didnt check myself), they released unfinished product, with promise of 300+ functional apps on the product or something like that. Would anyone consider buying that thing in any universe from someone who has history of promising stuff he cant deliver? I mean probably there would still be someone who buys that, but you get the point. Again, just my best guess.

[@jesuskock]

This is one of my favorite channels in general

[@gwnyc10009]

Coffee i totally feel ur pain brother. In a similar vein, I work in STEM and (for the most part) every time I have to interact with one of these smarmy staffing agency types or the like, it makes my skin crawl. If they were a nation and there was an Olympic event whereby the goal was to speak the absolute least amount of meaning via employing the most voluminous amount of words (*bonus points for flattering you and/or pointing out how much value they are bringing to the table) they would collect gold medals like…-a lot.

[@Aye_Nyne]

We’re watching Coffeezilla go down the Rabbit hole

[@Steverz32]

VoidZilla/CoffeeZilla never misses a shot he takes at scammers! 🎯😂

[@uuu12343]

"I'm right, you wrong, you stuuuuuuupid, bla bla bla bla bla bla bla bla bla bla I CANT HEAAAAAAAR YOU" - Rabbit

[@uuu12343]

Wow, ngl, I couldnt last halfway through Not because of the video, this is amazing stuff, but the PR and the company's responses triggered some PTSD in me and my hate for HR and company responses This is the worst of the worst, the scum of scums

[@snestah]

I am all for this channel becoming Coffeezilla: After Dark

[@Charles-dg2gk]

Extra dark

[@harukakonoha2126]

It is good that you making videos about this so people know which part where u and their team that didn't agree something than they saying later on that part 2 video now i understand n ur viewer understand 🤣🤣

[@0xPHN]

get the coffee machine ready fellas

[@ujtb]

I still consider the Rabbit R1 as a promise of an awesome device which I ordered, but after getting to know the company as it appears in your email trails, I've cancelled the order. So thank, for showing the colours of such company. The CEO should have indeed addressed the concerns with you directly. Cheerio.

[@LordTails]

I have a Master's and had a focus in data analytics (meaning lots of machine learning, AI, etc. type of scripting using R and Python). Full disclosure, LAM was not really mentioned in my classes if at all due to how fast AI has been evolving recently so i had to look up a lot on how it works compared to LLMs. However, based on everything I understand of Rabbit and how it works, it is technically a "LAM" in that it takes information and converts it to actions BUT it does so in a very limited and poor way, if you even consider it a LAM (based on my best understanding of the LAM proper, the R1 does not hit those metrics considering how effectively it does but let's assume it has a LAM for the sake of argument). The goal of a LAM is to get it so I can point a camera (or sensor) at an object or say something, the model breaks it down, and does particular actions (ie. point a camera at a flower in Home Depot and lets me order it or nutrients it needs, say I want to order a pizza and then orders it as I like it). While the R1 meets those metrics in an extremely limited sense for what few things it actually does, it does not do so well. Doing calls to GPT, depending on Playwright, etc. is a great way to have a system that is not very scalable or resilient. One UI change or a change in the API call that is expected, and suddenly the product does not work. If you wanted to use a browser-based automation, you'd need something that can correctly catch itself with changes, and react accordingly. The technology for that type of browser automation is there but, from my understanding, not with Playwright (I am more partial to Scrapy for open-source browser automation so my knowledge of Playwright is extremely limited but do know it ain't easy in Python considering you want it to be accurate but fast). Heck oftentimes it is better to do calls to site APIs for automation where possible just because of how volatile frontends managed by other companies can be usually backends at least will change much more slowly if at all. This is a problem for any automated tool, even the best ones. The R1 feels more like a rough proof-of-concept, and not something I'd be willing to pay a dime for. If this was a kickstarter, I think the dialogue would be very different, and maybe then I could be convinced to part with several bucks just to see where it goes. Call me crazy but I do not invest in promises of a halfway useful project in the next several years, especially when the company in question has gotten in over their heads on what are ultimately several incomplete projects [stages]. TLDR: someone who studied machine learning academically says the R1 is TECHNICALLY (the word does some very heavy lifting) a Language Action Model-like device, but a POC at best. I feel like in a short amount of time someone is going to make an open source LAM proof-of-concept that is more feature-rich. If I am proven wrong on it or any of this, I will happily eat my words on it and not gaslight anyone.

[@AD173]

Did you read their patent? If not, you should; It explicitly states that they are reliant on LLMs. Per their patent, they are not going to use APIs since one sub-claim in their main independent claim is the translation of requests into "user interface (UI) actions". Regarding open-source LAM alternative... It already exists; Meta released Toolformer last year, and it uses APIs.

[@LordTails]

@@AD173 so at the point not an LAM per their design. Like I said, I put for the sake of argument that it is but does so poorly at best. With their patents sounds like an admission it isn't an LAM altogether. Did not know about Toolformer. Thanks for the info. EDIT: I will check out the patent and update my post accordingly for future reference when I get the time. Now I am curious.

[@kellycasperhanson4426]

🚂 Riding along with the Void on the "Rabbit Gaslight Express Train" was quite a thrill, even if I know zip-nada-zilch about the tech info. Keep the fun ride goin' with any more updates from the 🚂 Rabbit crew!

[@cohan88]

The LAM is the email responses you received along the way...

[@superman1251]

Coffee: is the sky blue? Rabbit: the ground is clearly purple

[@brucebaker810]

Please give us the name of the employee who made these spurious allegations about sky colour. We may, in an informal discussion, have discussed pallette options for the Firmament. But no final announcement was made committing to a particular colour choice. Please note that, while we sunsetted the project, sunsets now have a variety of non-blue shades. Although, to date, we have not added user controls for sky hue. Also note the 87 Quadrillion pixel horizon-to-horizon blue-green we have installed at great expense. You're welcome.

[@superman1251]

@@brucebaker810 ☠️ I’m sorry that I dont have the correct certifications on the basic primary colors. I am not qualified to know the sky color.

[@brucebaker810]

@@superman1251 well I know that Coffee is black. White. It's complicated.

[@mwwhited]

It blows my mind at some of the stuff that people will “invest” in.

[@Flyingjaffacake]

We need people like you digging into other people's dirt. Well done

[@lukaspiatek7586]

I think they had the R1 reply back to you each time that’s why the responses were so bad

[@snestah]

👏👏👏

[@vprwave]

Probably a R1 development cluster, that's why they signed with "Rabbit Team".

[@deeplearningexplained]

@voidzilla LAM is most likely a toolformer, not an LLM agent. It doesn't directly control the UIs that's hard as heck to pull out and prone to mistake. However, toolformers can go from receiving a prompt to making the right API calls, which is pretty cool. In this particular case their thing go from prompt to API calls for dumb playwright script. You can get a toolformer LLM by leveraging in-context learning to teach LLM how to use various APIs. It was shown in the original toolformer paper that doing so makes even smaller model much stronger than ones with order of magnitude more parameters. In their case most of their APIs are just duck-taped scripts they need to maintain over time. I don't want to speak for them, but the strategy seems to be to showcase that this process works well and then to work with companies like Doordash/Uber in order to have direct access to their API. If they managed to pull that off then a toolformer LLM + proper API access is the most solid way of building that type of voice assistant tech.

[@AD173]

Well, considering one of their employees said it doesn't exist, I think the most likely option is that it really doesn't. Even their patent is so broad, it doesn't really say much specific. Except, the patent does describe what voidzilla is describing and not something like a toolformer. Why would they use a toolformer when their patent literally contains these sub-claims among several others in their first independent claim: performing streaming large language model (LLM) or natural language processing on the speech recognized user input to generate a first output; communicating the first output to one or more agents; receiving a second output form the one or more agents; a generating a result from the second output from the one or more agents;

[@SeeBiscuit1]

I hate lies too 😕. I mean no one's perfect obviously, but some of these powerful people with tons of money and influence on people's lives are all liars. Sinners, even. BIG demons out here.

[@meggylee8078]

In summary, Coffee: This statement on your website is false. Rabbit: Please see our website for the facts. Coffee: But your website is the lie I'm asking you about. Rabbit: You clearly are not engaging in good faith.

[@masharubi]

they tried to gaslight you about their own gaslighting. the jokes write themselves

[@mklptrk]

You’re like patience on a gd monument dude, going back and forth with these clowns. Keep fighting the good fight!

[@KathyClysm]

Coffee spilling the tea 😂

[@DetectiveStablerSVU]

Your rabbits do this to??!; duuuuuuude. I KNEW i wasn't crazy

[@rebeccaspratling2865]

😂