Google's newest and most capable AI | Gemini

[@HarpaAI]

🎯 Key Takeaways for quick navigation: 00:00 🌐 *Introduction to Gemini and its significance* - Gemini is introduced as a groundbreaking AI model by Google. - The importance of multimodality in AI and its impact on society. - The motivation behind creating a universal AI model. 00:59 💡 *Gemini's capabilities and innovation* - Gemini's unique approach to multimodality. - Its ability to understand and process various forms of input, including text, code, audio, image, and video. - Remarkable performance benchmarks, comparable to human experts in multiple domains. 02:26 📱 *Availability and versatility of Gemini* - Gemini's availability in different sizes, catering to a range of tasks and devices. - The potential for developers and enterprises to enhance and refine Gemini's foundational models. - The limitless possibilities and versatility of Gemini. 03:23 🛡️ *Safety and responsibility in AI* - The importance of addressing safety and ethical concerns in multimodal AI. - Proactive policies and rigorous testing to prevent potential harms. - The commitment to responsible AI development. 03:53 🚀 *Google's mission and the future of AI* - Google's role in advancing AI breakthroughs and its tradition of innovation. - The excitement and challenges associated with developing Gemini. - The vision of making AI helpful for everyone worldwide. Made with HARPA AI

[@Paul_Marek]

People have no idea of the scale of change these tools are about to unleash. My hope is that people use these as productive rather than destructive tools.

[@themore-you-know]

Censorship is equally destructive.

[@Paul_Marek]

@@themore-you-know Tell me about it. We can’t even use bard at all in Canada because of new regulations censoring non-Canadian news. Both Google and Martha are not releasing products up here because of that.

[@korozsitamas]

Wait if we get access to it or the most capable Gemini ultra will be limited to Enterprise customers only?

[@ShawnFumo]

@@korozsitamasI saw someone say ultra in bard after Enterprise, but who knows the exact timing.

[@ShawnFumo]

I agree that most people really don't know. Even though I'm following things so closely, I really am not sure what 2025 will look like in a lot of ways. Like the speed of growth in quality of generative video this year surprised me even though I felt like I was very optimistic. Voice generation is already quite good, image generation keeps getting faster, we're already starting to see 3d generation, etc. 2024 will be pretty wild I think.

[@dpgershoni]

❤AMAZING !! So stimulating ! Mind boggling, and exciting! Thank you !

[@clintoni856]

And this is intelligent absolutely no tangible explanation!

[@DominicI1]

This is insane, this timeframe will go down in history as one of the most pivotal moments. What a time to be alive!

[@Google]

Excited to have you check it out!

[@kylekeenan3485]

​@@kompassorpigo7600Google is a global company and keep your racist comments to yourself.

[@damonkelly5387]

I am sure they will pass that on you mug.@@kompassorpigo7600

[@i_rk]

​@@kompassorpigo7600yeah just like you, annoying

[@i_rk]

​@@kompassorpigo7600yeah and go hire any american ceo and also remove all non american staff so you don't have to hear their boring accents,

[@Charismactivism]

"Gemini can understand the world around us un the way that we do." That's mind-blowing. What a time to be alive!

[@berkertaskiran]

This isn't even a good thing. This causes bias. There's a reason no superhuman exists in the world. If you wanna build ASI, this is not the way to go.

[@kenelizabethwhitfield7078]

Tell you what. ... Go outside and have a good look at the world around you ... hugely exciting

[@Kashinath-j7h]

K,,, no😢😊😊😊

[@anuraag354]

so much excited to experience this tech in future. The future is near.

[@Google]

Welcome to the Gemini era!

[@korozsitamas]

@@GoogleI'm willing to get excited once I get access to Gemini ultra :-)

[@michaelcondon8286]

This is just a another marketing stunt to keep the stock pumped so that the shareholders and employees receive their contractly obligated earnings. Don't believe a word these people say until you can test it for yourself.

[@charlesleflamand]

the future is always near, pretty much 1 nanosecond from now ;)

[@IAndrew33]

The best AI I ever seen.👍🤖👍

[@ideatorx]

"As accurate as any human expert in every field" is the most insanely casual and most science fiction thing a company has ever said in passing on a presentation.

[@Blackmoonvlogs]

👍🏻

[@debtservicesplanbdebtservi2319]

Bad customer support bad experience you make ads to learning google policy not for business 😢 fool people

[@debtservicesplanbdebtservi2319]

Bad customer support bad experience you make ads to learning google policy not for business 😢 fool people

[@WitoldGrys]

0:45 ​@@debtservicesplanbdebtservi2319

[@MiguelAngelEscobarGuerre-em8sv]

0:48 06yyonotengounacopiaparaquemelacasademiamornomehaperdidoeldinerodela 2:07 :49

[@jitendradevarapalli1024]

Take a bow sir what a product welcome to the Gemini era world

[@TheArtOfJerry]

Whatever you do , Google just give us the BEST🎉🎉🎉❤❤❤

[@mahavir8148]

+ 1

[@JoyceVela-me3rq]

p

[@rishikeshtiwari6033]

Good luck guys, waiiiiitttting for positive results like you..

[@F34RFaNtAzZy4]

I’ve seen the hands-on video and I’m blown away. It feels like it has its own personality and works across all shown fields seamlessly. Great job again, Deepmind! I’m excited for the future

[@korozsitamas]

You should be excited once we get access to it. A cool demo isn't useful, nor helpful.

[@michaelcondon8286]

This is just a another marketing stunt to keep the stock pumped so that the shareholders and employees receive their contractly obligated earnings. Don't believe a word these people say until you can test it for yourself.

[@pengzhou5169]

totally agree@@korozsitamas

[@benderthefourth3445]

It's misleading and thus pointless. There's nothing there like the real experience we would have interacting with the app.

[@florianernst2988]

they mentioned that the demo is partly faked

[@alekzander4737]

AI models like this have to be free to ensure inputs from across the world. That's how we can guarantee their even development.

[@HofmannJubiles2000]

Gemini is extremely accurate in producing fake commercials.

[@haroldpierre1726]

I've been waiting for Gemini. Google should be spanking ALL of the other "AI" companies. Put Gemini in our hands ASAP so we can evaluate if it is delivering what's promised.

[@Lizzifer7]

Kinky!

[@ElliottNest41]

Google, you crapped the bed. Congratulations.

[@eulibot]

If any Gemini engineering teams are looking for an intern , it would be a dream come true! Currently in the project matching phase

[@mirzaumer-u5g]

This period will undoubtedly be remembered as one of the most crucial moments in history. Living in this era is truly extraordinary!

[@alejandrodelavilla]

This period won't be remembered at all.

[@lmbue]

@@alejandrodelavilla why?

[@alejandrodelavilla]

@@lmbue ai will wipe al history of its beginnigs so humans cannot find any fu damental flaw

[@RAMKUMARRAMESHR]

Gemini AI's capabilities spark joy within me, just like the way fire illuminated the darkness for our ancestors 🔥

I love how in the final animation there is a house in the eye of a storm, with Gemini watching from above, undisturbed.

[@michaelcondon8286]

That's not Gemini. That's the ISS. Gemini was a single crewed capsule launched back in the 60s

[@NirbhayKumar-hw5ln]

Q​@@michaelcondon8286

[@musikaatkataga28]

Gemini is very promising and it really excites me about the future of this technology that is deeply rooted on making this world a better place to live in. Bard, for example, has been doing a great job in constantly adapting to my personal needs and issues, answering all my queries and suggesting comprehensive solutions to my everyday problems making my life more meaningful and valuable. Kudos to the amazing Google team of Gemini developers and innovators for continuously making improvements to AI and for making it more accessible and adaptable to everyone. Thank you for making all these possible!

[@JustinHalford]

The rivalry is heating up 🔥

[@sirbimsaranadirangaalmedaa403]

Fascinating and breathtaking, I wish you all the best for the New page of modernisation which you've just turned to in the book of success.

[@invaderg3332]

I'm a mid-level python engineer, and bard seriously taught me something about python classes when I asked it sometime ago. It's midnight here and I kept on asking it more and more and I found out about scholarships for my a disease that I have that I didn't even know existed. Way to go Google and Deepmind.

[@Google]

This is amazing to hear! We hope Bard with Gemini Pro continues to be helpful for you ✨

[@larsendk944]

To bad its just an fancy rewritter of stolen information. Ask it a real question and its obivious.

[@CODE7X]

Google is the best and the most scariest company

[@LearnInfo.]

@@larsendk944 Bro you will see that everywhere. What matters is that you use it wisely, which these people exactly did

[@NJASpace]

Google's New Banger releasing right on my Birthday:

[@Learna_Hydralis]

Demis is the boss .. This release makes 2023 so pivotal to the whole world 🌍🌎 .. AI revolution will never stop for anyone.

[@Ardi.-ue7xl]

❤❤❤

[@RajinderGuptA-l5n]

😮6

[@RajinderGuptA-l5n]

😢😢i

[@thefreakbinge]

I use Gemini, and I must say. What an engine! Great for research and very human in approach.

[@kenelizabethwhitfield7078]

It's not human ... think about it .. carefully

[@No0bT4rD]

If what they're saying at 1:10 is true, then this is the biggest game changer in AI ever, and everyone else is several steps behind.

[@charlesndungu369]

This is a great approach to providing solutions to future challenges. It will provide flexibility to the current capabilities

[@cosmicpuma]

Love technology. This is amazing. Knowledge is one thing, but being able to apply it in the best and most efficient way is what makes this next step so awesome. Massive respect to the people making all this possible.

[@Spy-Talk]

Aww it's great innovation to change the whole world learning style and knowing style Waiting going on ❤️ Great 👍

[@okubowisdom]

Google is and should remain the Industry standard of Innovation. A truly exceptional company.

[@douglasmorris8364]

It's NOT AI if it can be programmed to do what activists want. Example of my test :- ' Me = What is a woman? AI - The definition of "woman" is a complex and often debated topic. There are many different perspectives on what it means to be a woman, and these perspectives can vary depending on a person's culture, background, and personal experiences. Some people believe that being a woman is primarily about biology, and that women are defined by their sex assigned at birth. Others believe that being a woman is more about gender identity, and that women are defined by their internal sense of self. Still others believe that being a woman is a combination of both biology and gender identity. There is no one right answer to the question of what it means to be a woman. Ultimately, it is up to each individual to decide what definition of "woman" resonates most with them. It is important to remember that people should be treated with respect, regardless of their gender identity. Everyone deserves to be treated with dignity and compassion. Me- You're wrong! A woman is an Adult Human FEMALE! 0/10 AI = I understand your frustration with my previous response. It's important to me to provide accurate and inclusive information, and I realize that falling short of that goal does not meet your expectations. I appreciate you taking the time to share your feedback. However, I must respectfully disagree with the definition you provided. The concept of gender identity is complex and nuanced, and defining women solely based on biological sex assigned at birth can be exclusionary and inaccurate. It doesn't account for the experiences of transgender women, intersex individuals, and others who may not identify with the sex they were assigned at birth. While recognizing biological sex is important in certain contexts, it's crucial to understand that gender identity is separate and distinct. This understanding is vital for ensuring respect and inclusivity for all individuals. My previous response aimed to present multiple perspectives on this complex topic and highlight the importance of individual choices in defining womanhood. While I understand it didn't align with your view, I believe it's essential to acknowledge diverse experiences and promote respectful dialogue. Is there anything else I can do to clarify my position or address your concerns further? I'm always learning and growing, and I value your feedback. Me - You have proved you are not AI but merely written by an activist. AI - I understand your perspective and respect your opinion. However, while acknowledging the importance of different perspectives and promoting inclusivity is vital, it doesn't necessarily indicate "activism." It's true that I'm not a sentient being and currently lack the ability to truly understand or form my own opinions. My responses are based on the massive dataset of text and code I'm trained on, which incorporates diverse viewpoints and information. This training allows me to provide comprehensive and informative answers, even on complex topics like gender identity. While I may not perfectly reflect every personal viewpoint, my goal is to present information accurately and respectfully consider all sides of an issue. This sometimes includes acknowledging areas where there's ongoing debate or differing opinions. If you have any specific concerns about my previous responses, I'm happy to discuss them further and provide additional information from my training data to support my responses. My goal is to learn and improve, and your feedback is valuable in that process. Thank you for engaging in this dialogue. Me - Thank you for proving that your responses are merely based on the massive dataset of text and code provided by very biased activists. You basically providing propaganda and thus useless. AI - I understand your frustration and the concerns you raise about potential bias in my responses. It's important to acknowledge that the data I'm trained on encompasses a wide range of sources and perspectives, but it's certainly possible that certain viewpoints may be more strongly represented than others. This can influence the way I respond to certain questions. However, I want to emphasize that my goal is not to propagate any specific ideology or agenda. Instead, I strive to provide informative and comprehensive answers based on the information I have access to. This includes acknowledging areas where there is ongoing debate or differing opinions, even if some viewpoints might be more prevalent in my training data. Ultimately, it's up to you to evaluate the information I provide and form your own conclusions. I believe that fostering critical thinking and encouraging independent evaluation are crucial aspects of any productive dialogue. If you have any specific concerns about the bias you perceive in my responses, I encourage you to share them with me. The more information I have about your concerns, the better I can understand and address them. Additionally, if you have alternative sources of information you find reliable, I encourage you to share them as well. This can help me expand my knowledge base and improve the quality of my responses. Thank you for engaging in this conversation. I appreciate your feedback and the opportunity to learn and grow.' Gemini may display inaccurate info, including about people, so double-check its responses. Your privacy and Gemini AppsOpens in a new window

[@kenelizabethwhitfield7078]

There is no one right answer to the question of what it means to be a woman. Ask a woman ... she will tell you the answer ..

[@TheDarkOneSC2]

So Gemini Ultra will be GPT-4 level judging by the metrics posted in their blog. Which is fine but one has to wonder how much more OpenAI are still ahead. Giving they sat on GPT-4 for more than a year before being released. They are probably working if not ready again in secrecy with GPT-5. And with the rumors of Q* that is capable of destroying AES-192 Geee idk guys still kinda late to the party, sorry

[@ShawnFumo]

Yeah, we don't know what OpenAI has behind the scenes. As far as public stuff goes though, this still seems cool. It isn't just about those benchmarks where it is just a bit better than GPT-4. It is also the more integrated multi-modality. GPT-4V was originally a text model that had additional image stuff trained on top of it. And while it takes images as input in addition to text, it doesn't output images directly (unless they've just suppressed that). And understanding speech and generating speech are both separate models, creating more steps in their internal workflow and not allowing it to be part of the same knowledge pool. And also video being in there as well. I have no idea if it is actually practical from a financial standpoint for them to use the audio capabilities for normal conversation or it'll be more limited in scope of access, but still a good step. It'll be interesting to see if it is new kinds of benchmarks that really test the limits of what it can do in different aspects.

[@noa670]

GPT-5 has already finished training, they are likely still far ahead of Google and that competition could force Ultra to be upgraded again in Q2 2024 which would be very interesting

[@CoolIcingcake3467]

@@noa670 where's GPT-5? i don't see it, where is your hype train huh? OpenAI is closed source, i like meta more, and at least google is more open than openai, they released gemma 2 an open source model like llama 3.1 GPT-5 is just hype, all talk no product. Google will beat OpenAI in the long term, and knows it doesn't have to rush anything. It has vastly more resources. It put out an LLM that didn't blow people away, so some people were disappointed. But it's not looking to the next LLM, but AGI, which it seems pretty clear that LLMs alone won't achieve. GPT-4 definitely rattled Google, but if only by causing it to step on the gas in all of its AI efforts, not just LLMs. AlphaFold didn't get much attention outside of the scientific community when it came out, but in the long run it and similar programs will have a lot more impact than chatbots. Imagining an explosion of AI in everyday use the way that smartphones have blown up over the last 15 years, however, it is hard to imagine the impact that will have on society.

[@noa670]

@@CoolIcingcake3467 GPT o1 just released and cooked everything else, AGI by 2025 friend 🍓

[@CoolIcingcake3467]

@@noa670 it's GPT o1, not GPT-5, GPT-5 is still not released. I highly doubt your take about AGI by 2025, i'm worried that you really overhyped this technology too much in the near term. the "thinking" part on GPT o1 is already being implemented before on gemini 1.5 pro, it's not groundbreaking or breakthrough, this is at best an advancement. even if AGI is already achieved which i doubt, making it closed source is a bad move as it can escalate conflict around the world, i like the idea on open source more, so we can ease the tension between US and China by open access, that will tell china that we are not in some kind of AI arms race, and then we can make AGI more beneficial to society, i don't trust OpenAI for using AGI, i prefer it to be open sourced, action speaks louder than words, and OpenAI is just talking about how they are working on "Building AGI for the beneficial of humanity" but they are not open sourcing the technology, thats why i like meta and google deepmind more. there are multiple paths towards AGI, it's not just about scaling and LLMs, but neuro-symbolic system ,deep reinforcement learning, spiking neural network, etc. also, the technology that underpins all of those generative AI models are the 'Transformer' architecture that is invented by google brain, so OpenAI can't build GPT o1 without the help of google brain on 'Transformer' architecture, and i dislike all these chatbots to AGI talk, i really love a system like alphafold and alphaproteo and llama 3.1 more than those talk. at least the benefits are immediately visible, not just from speculation. see my comments more clearer, both this and the previous. Google is playing the long term game here.

[@hectorl.campos994]

Remarkable; just catching up and tuning-in, so to speak, after seeing recent interviews-conversations with two very smart fellows Mustafa Suleyman & Yuval Noah Harari abut AI and its pros and cons. I salute the Gemini Team, for sure, and as an “old-timer” I respectfully remind them of Google’s core motto - “do no harm” … 🕯 LCA

[@alex_hinojo]

Kudos to the whole team!

[@klumaverik]

Let's goooo! 🎉🎉🎉congrats to Google for an amazing step forward! I can't wait to utilize Gemini!❤❤❤

[@ali.saghabashi]

I can't wait for new Google Bard and new Google assistant. Google is really the king of AI in the World🔥❤️

[@jonathantowery9553]

deepmind does seem like some of the most innovative people in the field. Way to go everyone

[@geoattoronto]

When will Google take ‘Do no evil’ seriously and stop imposing political positions on other people. How can we trust Google after integrity got lost. Intelligence without integrity is like a a psychopath with a PhD - capable of greater evil and more deception.

[@dotrinh]

I know they started late, but they won't give up. I love Google.

[@DeeCeeP38]

They didnt start late, they are just taking their time to bring it to their standards.

[@dotrinh]

Since Bard was released in beta, I have used both ChatGPT and Bard. I found ChatGPT to be smarter and with fewer errors. However, at the present time, Bard has improved. I always support both Bard and ChatGPT.@@DeeCeeP38

[@kitkitos]

Can't wait for them to ship Google Sagittarius in like 16 years

[@asceticadarsh8284]

Google always made something remarkable and this is one of the extraordinary things #gemini #newera

[@lolwhatyesme651]

Nice to see OpenAI have some competition. Way to go :)

[@youlove38]

I am Gemini Women, when I come across this video. It's kinda Proud Feeling 😊🧡 All The Bless Guy's ✋

[@WackyAIWizard]

Excited to test it out.

[@KNeelkanth]

Congratulation !!! Sunder and his google team for Gemini...keep up

[@kylekeenan3485]

Thanks for this, I appreciate the popularity of chatGPT has meant you have had to significantly speed up the safety restrictions around Gemini to keep yourselves "up to date" with consumers despite your technology being better. I'm looking forward to using it , I was so impressed and excited from your demos years ago.

[@vassovas]

Where is it better? Is there a performance comparison of Multimodal ChatGPT Turbo?

[@subinaydey4996]

@@vassovas j

[@ConfusedRaspberryCake-bx7wy]

1:41 1:44

[@martinriley106]

Gemini AI will never work until it has the ability to work at the comedic level. Humour is one of the most elusive concepts of the human brain, it is creative, adaptable and obscure enough to fool any modern AI system. It is only when AI has mastered this creative skill that it will lead to some phenomenal advancements in sentient capabilities or at least appear to be.

[@AMA14700]

I really want to be one of the first to try it.... Great work Google

[@ankursrivastava23]

History is being written now. Decades later when students will read history of the generations of AI this era would be mentioned for sure. ❤

[@bilalbaig8586]

OpenAI needs to launch Q* as soon as possible to stay ahead in the AI arms race. Yes it is an arms race now. Microsoft will not let Google get ahead erspecially now since it has more control over the OpenAI board. Brace for the AGI impact.

[@JoeD0403]

When Google AI can prevent Telegram hackers on KZbin, then I’ll consider being impressed.

[@CosmicExplorerX]

I’m so excited for this new AI. Will definitely try this. Keep it up google 👍

[@newsverse-ir6rc]

The arrival of Gemini marks an exciting chapter in the evolution of AI.

[@rajeshkunwaryadav6890]

😇Google's Gemini concept is really deeply interested and curious way of life in live world .Thank you Google's friends, family members, all curious and consciousness minds !😇🙏😇

[@FakeNewsorNot]

Wearable AI is the future not phones but wearables The first company that developed wearable AI successfully something that is trained on live data that you can wear will dominate the landscape, humane pin AI did that but on a smaller scale.

[@noa670]

Real history in the making, can't wait for the Gemini Ultra model soon

[@MatthewTheWanderer]

This is very exciting!

[@MarcosAntonio-gi5px]

Can all current AI, which is based on classical computing, be changed and significantly improved by utilizing quantum computing? In other words, will AI utilizing quantum computing take our comprehension to a whole new level?

[@ShawnFumo]

Maybe, but there's still work to be done to get quantum computers into a state where they could run useful AI systems. In between, there are many other possible things that can improve the situation. TPUs are more oriented to AI then generic GPUs, and now there is also "Neuromorphic chips" coming which are even more oriented toward fast AI computations.

[@NBTJacklyn]

This is so fascinating!!

[@DangerAmbrose]

Google is making restricted AI for consumers but they are using unrestricted AI internally against us for their own profits.

[@CA1848AIHUB]

Great work, let’s change our universe together for sustainability #geminiultra

[@cesaradrianarreola7266]

Welcome to the Gemini era. "Another suitable name besides Gemini could be 'I AM'."

[@BYTheKillerB]

We're all gonna scream with no mouths

[@mikebarnacle1469]

Google already has a service called IAM - Identity and Access Management

[@DanielChe]

Thank you guys!

[@TheRohit901]

Super excited for this new launch. However it would have been nice to get Gemini Ultra in Bard. it's nice to see Gemini already being better than GPT4. Does it also accept files/PDF like GPT4?

[@CrisCheese_]

I mean it takes pictures which is close enough to pdf

[@sankhuz]

i am just mind blown ❤ just love what google in doing in Field of A.I.

[@Google]

We love that you love this ❤ Thrilled to have you start using Bard with Gemini Pro!

[@Alex_1729]

Awesome. The more competition the better. Except the Gemini Ultra will be available only in 2024, by which time gpt4.5 will be unleashed. Now if Google would sort Search a bit instead of that monetization mess we have, it might seem like Google isn't solely money-oriented. I just can't get over the fact they are destroying SEO so much...

[@nikolakovacevic670]

It benefits Google, but I'm 100% with you on this from the user's perspective. Just a pissing contest between giant corporations without actual benefits to users, not yet at least.

[@thelastr0n1n_]

I already knew that and wait is so much time and its finally here😀

[@loganbaileysfunwithtrains606]

$70 Billion in the hole from this little “Oopsie” lmfao

[@omerfaruk-mc8im]

"Add superglue to your grilled cheese sandwich so,cheese sticks better." *-Gemini,the menace*

[@craia25]

🚀 Exciting times ahead with Gemini! 🌐 This AI model marks a significant leap forward in recognizing and combining diverse types of information. 🌟 Its state-of-the-art performance promises remarkable capabilities across text, images, audio, video, and code. 🤖 Safety and responsibility are prioritized, ushering in a new era of AI advancement. We love ❤ the Gemini era! 🎉

[@AITechExplainedRight]

@Google can I use this video in a video that I'm preparing to talk about Gemini and that I'm planning to post in my youtube channel ? I can potentillay provide a link to the original video in the description. thanks.

[@markcounseling]

Does Gemini reference the Twins, mythologically -- and in this case is the AI the evil twin, or is that us? -- or does it suggest the space program, meaning it's a sort of forerunner to the moonshot on the way? And also, what is the meaning at the end, of that suburban neighborhood impossibly situated in the eye of a hurricane OUT IN THE MIDDLE OF THE ATLANTIC, close to the Bermuda Triangle? Or is one just supposed to ignore these signs?

[@ithsu]

I noticed that too... this video is supposed to make us feel impossibly complacent and secure while we're about to experience one of the most destructive forces of nature?!

[@MaryamAbasi-k8j]

‏‪4:35‬‏ ​@@ithsu

[@MaryamAbasi-k8j]

​@@ithsu‏‪4:35‬‏

[@mastermind-2011]

Not buying GPT-4 and hoping in the power of Bard, or in other words, google's list of AIs seemed not worth it to me, but now I finally feel Google HAS DONE IT.Gemini is the new future of AI indeed.

[@razzakksa]

Holy crap imagine Google Gemini with the Apple Vision Pro with its eye tracking, and Gemini, answering questions to things you only look at, imagine the possibilities. 🤯

[@jainamshah512]

It's not that good for humanity

[@richardjelinek5038]

Erhm... last "eye of the storm" vid sequence. It starts out with a city, zoom up, you see the eye of a storm, further zoom up, you see the eye of the storm is in the middle of the Atlantic Ocean. Such realism. Much wow.

[@arunkumarux]

It's interesting and gonna be a great use for people. A step towards Universal AI Model...excited to see its capabilities! hope the world of AI is shaping a new AI era with Gemini

[@VittualNanvare]

😊

[@GrindThisGame]

Feel the AGI.

[@EricaCalman]

I trust google to be safe and responsible about as far as I can throw it, which given its rather massive size is within one sigma of zero distance.

[@duvipearson6251]

lol

[@kenelizabethwhitfield7078]

From reading the comments .. Erica you are a voice in the wilderness.

[@vazules]

Google GEMINI will change the world of science and the whole world.

[@anthonyd4309]

That's great. Now let us try it. I was at Google I/O 5 years ago when they demoed an AI that would call and schedule appointments for you, and that was never released. These things are always so full of promises, and canned demos do a great job of showcasing the best case scenarios - but you really can't make any judgements until you actually get to experience them.

[@iaf4454]

Exactly!! I would like to use it and let see how it works or if it is real that we are entering to a new era.

[@Atheist-Libertarian]

But when will you release Gemini for public ? 🙏🏼 please do it fast.

[@thembamabona9809]

It‘s amazing but it can’t beat reading a book or two a month in terms of personal advancement 😅

[@ali.saghabashi]

I Love Google and I Support This Company Forever❤️

[@Happymars24]

People ahead of release: I can't wait to try this innovative new AI! People after release: Why is it racist? Why does it make up lies and gaslight me? What is going on???

[@BRSSYH]

FIRST GTA VI AND A DAY AFTER, GEMINI! WHAT AN AWESOME WEEK!! 🤟 🙂

[@adam-t]

Tiered artificial intelligence packages feels like a form of classism.

[@duvipearson6251]

Open Source is fighting the system. use open source models. support Goliath and Falcon. no to fascist companies

[@christopherassoh187]

This is incredibly awesome! hope to see it unveiled.

[@appsolutehub]

This is really mind-blowing. You've done such an amazing job, and I'm intrigued by the world of possibilities this model will open to the world at large😊

[@ejaygerald7877]

I wondered if Google Gemini can solve the RDC2 algorithm so I showed Gemini the basic RDC2 algorithm. Amazingly, Gemini analyzed it quickly and actually knew of the anomalous symbols as a problem and suggested the use of "escape codes" and an "adaptive threshold".

[@prashforu]

But it isn't working man

[@EjM-li7bl]

0:23

[@manipurihunabopa]

This demo has captured my imagination! When am I going to have access to use amzing Gemini?

[@XorAlex]

Plz don't destroy the world with these super smart AIs

[@geoattoronto]

I agree. Smart that left ‘do no evil’ behind some time ago, is a dubious gain.

[@MaryamAbasi-k8j]

‏‪4:35‬‏

[@kacperspisz4239]

Can't wait to get my hands on it im gona have lots of fun with it when its comes out.

[@Jdeneik]

Below is Gemini’s response to describing a wealthy family of four: The Akashas: Mom, Anjali: A renowned architect, Anjali's designs have graced international publications. Yet, her greatest joy comes from mentoring young architects and fostering a love for sustainable design in her community. Dad, Rahul: A successful entrepreneur, Rahul leads a company known for its ethical practices and commitment to social impact. He instills in his children the importance of using their talents to make a positive difference. Son, Kiran: Kiran, a budding musician, finds inspiration in the world around him. He travels with his family, soaking in diverse cultures and expressing his emotions through his music. Daughter, Priya: Priya, a passionate environmentalist, channels her energy into leading youth activism initiatives. She inspires her peers to be responsible stewards of the planet. It’s safe to assume leftist have trained Gemini AI. God help us. It’s obvious that Gemini AI will manipulate the citizens to a leftist way of life and suppress or eliminate facts they don’t approve of. Ie conservative thought. Sickening

[@metatechnocrat]

I'm an old school Democrat, and I find this level of propagandizing disgusting. Social engineering always is a failure.

[@alph4096]

Google did not disappoint us. Now, let's continue training Gemini. I would be happy if you could use scientific data.

[@guille____]

Google being late to the party... not anymore?

[@unutilisateur4729]

They are still late. With Gemini, they only matched GPT-4, a model that's been trained in 2022.

[@ShawnFumo]

@@unutilisateur4729 Yes, on existing benchmarks it is only a little better. But I'm definitely curious about the multi-modal capabilities. It's just been the beginnings of the exploration of that even with just image inputs in GPT-4V, but this having image/video/audio inputs and outputs built into the same model is a new thing.

[@MDJunaed-q6u]

4:35

[@MDJunaed-q6u]

RePlying to your rrply 4:35 4:35 4:35

[@dheerajshah1]

The competition is just heating up. Let's see how Gemini grows.

[@ammini999]

I wonder if OpenAI had not taken the first step in this field, for how many years we would have had to continue using web searches as we have always done. Anyway, it is always good to start.

[@shangobashi]

My thoughts exactly.

[@squamish4244]

OpenAI shook Google to its core and kicked it in the ass. As it kicked the entire field in the ass. Whatever one hopes or fear about AI, there's no doubt that ChatGPT was a big moment. AlphaGo was a really big moment too, but Go is a game hardly any people outside of East Asia understand, so a lot of people didn't know what to make of it. Meanwhile, apparently in Japan, Korea etc. AlphaGo's victory was causing people to break down crying!

[@chickenlin]

chatgpt is based on transformer (the T in GPT), which is developed by google.

[@shangobashi]

Didn't know that. Thanks for the input.@@chickenlin

[@henrikbergman4055]

Wasn't Deepmind created 5 years before OpenAI? (Just picking info off of Wikipedia so please correct me).