It turns out that fancy video Google made to show off its new “AI” was… Well, not “faked”, but definitely a bit staged.
Google also admits that the video is edited. “For the purposes of this demo, latency has been reduced and Gemini outputs have been shortened for brevity,” it states in its YouTube description. This means the time it took for each response was actually longer than in the video.
In reality, the demo also wasn’t carried out in real time or in voice. When asked about the video by Bloomberg Opinion, a Google spokesperson said it was made by “using still image frames from the footage, and prompting via text,” and they pointed to a site showing how others could interact with Gemini with photos of their hands, or of drawings or other objects. In other words, the voice in the demo was reading out human-made prompts they’d made to Gemini, and showing them still images. That’s quite different from what Google seemed to be suggesting: that a person could have a smooth voice conversation with Gemini as it watched and responded in real time to the world around it.
Parmy Olson for Bloomberg
Companies always lie. It’s in their nature.
Well that was a shitty thing to do. I was really impressed by the demo, but it was a total deception. Who authorized this lying? Everyone wants to see where the technology actually stands, but don’t lie about it. That damages your reputation.
Alfman,
Technically they had some disclaimers, and fineprint on the presentation. But they forgot who their target audience was. If this was an inspirational ad on what the platform could potentially do (and on a good day, with favorable conditions), yes would be a good presentation.
But if your target audience is current tech savvy generation, who are already using GPT or an existing open source alternative (like LLama 2), they would of course pick everything apart, and then try the demo queries on your platform.
They had one chance, and they missed. They can of course recover, but this will probably considered an “own goal”.
Well, I watched the presentation and didn’t see any. The video itself lied. While I see it now under a tiny “…more” link, those of us who aren’t explicitly looking for a disclaimer would never see it. One shouldn’t have to look for disclaimers to know whether the video content is honest. It should be honest to begin with. Meh, just my 2c.
I was genuinely impressed with the demo but feel conned by the way it was falsely presented.
Apparently they tried to pull off something like the original iphone demo video. But man… This was really stupid. Even more stupid than Bard.
The landscape of publicity available LLM services becomes increasingly murky, with companies pitching their “top tier” models to the press only to make their heavily stripped down versions with really unknown capabilities available to the public.
Moreover it seems that initially announced services get gradually dumbed down as public attention switches to something else.
All in all expect a lot of time saved and a lot of time wasted on AI version of bait and switch . As usual the attention and time of is used to influence the choices of the few who are the real customers.