- AI For Humans - The Newsletter
- Posts
- Say Hello To The (New) Future
Say Hello To The (New) Future
The new ChatGPT Advanced Voice is so good it has us re-thinking where AI interfaces are headed
Today on AI For Humans The Newsletter!
We’re all freaking out about ChatGPT’s new advanced voice
Meta blows us away with a raft of announcements
And a titan of Hollywood turns to AI
Plus, our can’t-miss AI app of the week!
Welcome back to the AI For Humans Newsletter!
This week’s big story is the launch of ChatGPT Advanced Voice, now available in the mobile app to everyone on a paid plan.
Advanced Voice Mode (AVM) is the dawn of a new something, depending upon who you ask. Though we’ve been able to use our voice to interact with machines for years, it required some level of effort by the user to adapt their vocabulary, syntax and cadence to work with the technology, usually to achieve some limited result; set a timer, dim the lights, play “Cotton Eye Joe” on loop… with AVM, users are experiencing “natural” conversations, where an assistant is capable (and quick enough) to make reaching for your pocket to mash letters on a keyboard to make a request of intelligence seem laughably antiquated. To quote that kid in Back to the Future 2, “You have to use your hands? Like a baby’s toy?!"
Hey Siri, re-hydrate my product integration…
Certainly, text has its place, just as video and gestures will eventually have theirs; and in time all modalities of interaction will seem natural and capable.
But this is time for voice to shine! So naturally, we’re going to stomp on it a bit:
Establishing a voice chat still feels “high-friction”. You have to load an app, select chat, wait several seconds for a connection to be made, and then you’re finally living in 2043. Once voice natively built into devices, and always listening (proactive versus waiting to be summoned), user behaviors will change.
The guardrails are too high: often users are frustrated by a new technology’s limitations, but here, the frustration for many has stemmed from feeling the tech is far more capable than they’re being allowed to experience.
Access to real-time data and “agentic” functions will be a big unlock: it feels like you’re chatting with a super-capable intelligence, but they can’t tell you the weather, movie-times or order Chipotle? We know it’s coming, we just thought it would happen before the machines could laugh at dad-jokes or impersonate “Jamaican-Elvis”…
TL;DR: If you pay for ChatGPT Plus, go have a chat with the future. And if you don’t? Give it six months and play with an almost as adorable competitor’s offering…
-Kevin & Gavin
3 Things To Know
James Cameron Joins the Stability AI Board
Another week, another first for Hollywood as we see arguably the first high-profile partnership of a major Hollywood filmmaker with an AI startup. While it’s great PR for Stability, who has been in the headlines for the wrong reasons lately, Cameron expressed enthusiasm to explore “the intersection of generative AI and CGI” as the impetus for coming aboard. As we shared last week, Lionsgate just did a controversial deal with Runway that we suspect is a sign of things to come.
New AR Glaggles Teased by Meta
Zuckaissance aside, Mark Z’s still making a beeline for any reality but our own. Meta grabbed headlines last week with AR glasses so cool you can’t even buy them. They’re called Orion - possibly a jab at OpenAI’s next top model? They also announced Llama 3.2, two new 11B and 90B parameter multi-modal models. The new models were presented as almost a footnote after Mark showed off his new glasses.
Hold still while I examine your oats
More Drama at OpenAI
OpenAI loves to bury drama with juicy product releases, and Advanced Voice was no different. Reuters is reporting Sam Altman will remove non-profit control of OpenAI, taking it for-profit, and bestowing Altman with 7% of the company. As OpenAI closes in on the largest VC raise of all time at a valuation of $150B, this would value Altman’s new shares at over $10B. Several leaders left the company the day of the announcement, including CTO Mira Murati.
We 💛 This - Glif
What it is: A quick (and free) way to produce silly AI creations, and entire AI workflows. Whether it’s a children’s drawing of hot dog city, an 80s Fisher Price hot dog toy, or a medieval hot dog stand, there’s something for everyone.
Why we love it: Silliness aside, every workflow can be “remixed”, and you can view every step that goes into its creation. The models, the prompts, the LORAs, the settings that went into the full workflow. Or you can just make hot dog jokes.
Are you a creative or brand looking to go deeper with AI?
Three ways AI For Humans can help:
Join our community of collaborative creators on the AI4H Discord
Get exclusive access to all things AI4H on our Patreon
If you’re an org, consider booking Kevin & Gavin for your next event!