- Everyday AI
- Posts
- Sam Altman’s New Take on Superintelligence: What it means and are we that close?
Sam Altman’s New Take on Superintelligence: What it means and are we that close?
OpenAI unveils Advanced Voice mode, Gemini added to Workspace, Russia uses AI to interfere with US election and more!
Everyday AI
September 24, 2024
👉 Subscribe Here | 🗣 Hire Us To Speak | 🤝 Partner with Us | 🤖 Grow with GenAI
Outsmart The Future
Today in Everyday AI
8 minute read
🎙 Daily Podcast Episode: Is superintelligence much closer than we think? Why does it even matter? We break down what superintelligence is and its importance. Give it a listen.
🕵️♂️ Fresh Finds: Gemini’s impact in the workplace, OpenAI’s Twitter account gets hacked and Hollywood figures call for new AI bill in California. Read on for Fresh Finds.
🗞 Byte Sized Daily AI News: OpenAI unveils advanced voice models, Gemini added to Workspace as default and updates its Gemini 1.5 models and Russia uses AI for U.S. election interference. For that and more, read on for Byte Sized News.
🚀 AI In 5: Unlock the secrets of creating custom GPTs that will streamline your workflow and save you precious time. See it here
🧠 Learn & Leveraging AI: Wondering if superintelligence is legit or a myth? We break down everything you need to know. Keep reading for that!
↩️ Don’t miss out: Did you miss our last newsletter? We talked about OpenAI's academy for developers, Musk investing in his supercomputer and Cloudflare offering AI content monetization. Check it here!
Sam Altman’s New Take on Superintelligence: What it means and are we that close? 🧠
Are we really 'a few thousand' days from Superintelligence?
Also, what the heck does Superintelligence even mean?
We're breaking down the latest hot takes from Sam Altman and simplifying superintelligence.
Join the conversation and ask Jordan questions on AI here.
Also on the pod today:
• Sam Altman’s Influence on AI 🏋
• Definition and Discussion of AI Levels 🤖
• Major Historical Transitions and Their Relevance to AI 📜
It’ll be worth your 31 minutes:
Listen on our site:
Click to listen
Subscribe and listen on your favorite podcast platform
Listen on:
Here’s our favorite AI finds from across the web:
New AI Tool Spotlight – Polymet is an AI product designer, Magic Inspector is an AI web test automation platform and Nuvio provides AI-powered financial management.
AI compute — Intel made a big splash with its new AI accelerator, Gaudi 3.
Google – Google has released a slew of articles about Gemini’s impact in the workforce at its Gemini For Work event.
OpenAI – One of OpenAI’s X (Twitter) accounts was compromised for a Crypto scam.
Someone has gained access to the official OpenAI Newsroom X account.
This is now the 4th OpenAI X account to be compromised in the past 15 months.
— Smoke-away (@SmokeAwayyy)
10:31 PM • Sep 23, 2024
AI Audio - Spotify has launched its AI playlist builder in the US, Canada, Ireland and New Zealand.
AI in Media – Major Hollywood figures like Mark Hamill and J.J. Abrams are calling for Gov. Newsom to sign a new AI bill.
AI Models - Replicate has made its Flux code open source and is adding some updates.
We're open sourcing our Flux code.
The community loves FLUX.1 text-to-image models. To serve them, we've made a bunch of improvements internally: image-to-image mode, NSFW checkers, and most recently, a 40% speedup.
Now you can use that code too. For free.
(link in next tweet)
— Replicate (@replicate)
10:35 PM • Sep 23, 2024
AI in Medical – Novartis has struck a deal with Generate:Biomedicines to develop targets in various disease areas.
Read This – Forbes has released the 10 biggest AI trends to watch out for in 2025.
1. OpenAI Unveils Advanced Voice Mode for ChatGPT 🗣
OpenAI is expanding its Advanced Voice Mode (AVM) to paying customers in the ChatGPT Plus and Teams tiers, with access for Enterprise and Edu users coming next week. The feature now sports a blue animated sphere and introduces five new nature-inspired voices, upping the total to nine—just shy of Google’s offering. While the rollout enhances voice capabilities and includes improvements in accent recognition, notable absences like the Sky voice and the anticipated video features leave users eagerly waiting.
As AVM aims to make interactions with ChatGPT feel more natural, it promises to be a game-changer for professionals looking to leverage AI in their work, despite some lingering limitations.
2. Google Integrates Gemini AI into Workspace 💼
In a strategic move to enhance productivity, Google is making its Gemini AI a standard feature in Workspace Business, Enterprise, and Frontline plans starting in Q4 2024, eliminating the need for a separate add-on. This integration, which aligns with existing enterprise terms for services like Gmail and Docs, ensures user data privacy by not training the AI on organizational information.
Additionally, Workspace Administrators will soon have control over how long generated data is stored, bolstering security with new SOC and ISO certifications.
Google also updated its Gemini 1.5 Pro and 1.5 Flash models and showcased significant benchmark improvements, a 15% price reduction, and increased rate limits, enhancing scalability for businesses.
3. OpenAI Unveils Multilingual AI Dataset 🚀
OpenAI has announced the release of its Multilingual Massive Multitask Language Understanding (MMMLU) dataset, now available on Hugging Face. This comprehensive dataset evaluates language models across 14 languages, including often-overlooked ones like Swahili and Yoruba, ensuring higher accuracy thanks to professional human translators.
As businesses increasingly adopt AI solutions across diverse markets, this resource equips them to tackle language barriers and improve user experiences. OpenAI's initiative not only positions it as a pioneer in multilingual AI but also opens doors for developers in emerging markets, potentially reshaping the industry's landscape.
4. Microsoft debuts New AI Security Measures 🔐
Microsoft has announced new capabilities aimed at improving the security, safety, and privacy of AI systems. The company’s Secure Future Initiative (SFI) now features proactive risk assessments in Azure AI Studio and transparency tools in Microsoft 365 Copilot, ensuring that organizations can confidently leverage AI technology.
With investments in confidential computing and innovative safety features, Microsoft is setting a high bar for responsible AI deployment across various industries. As businesses increasingly adopt AI solutions, these enhancements may empower professionals to harness the technology while maintaining data integrity and trust.
5. Russia's AI Strategy in US Election Interference 🗳️
A US intelligence official has revealed that Russia is generating AI-produced content to bolster Donald Trump's presidential campaign while crediting Vice Kamala Harris in the lead-up to the November 5 election.
This revelation is part of a broader effort by Moscow to employ artificial intelligence in shaping narratives and influencing voters, with an emphasis on conspiratorial themes aimed at the Democratic Party.
6. James Cameron Joins Stability AI Board 🤯
Filmmaker James Cameron is now on the board of Stability AI, the company behind the popular Stable Diffusion model. Cameron believes that the fusion of generative AI and CGI is set to revolutionize storytelling in film, as he aims to push creative boundaries further than ever before. Stability AI's CEO, Prem Akkaraju, views Cameron’s involvement as a pivotal moment for the company and the broader AI landscape, promising to transform visual media for the next century.
As Hollywood grapples with the implications of AI technology, this collaboration signals a significant shift in how filmmakers might harness these tools to enhance their creative processes.
7. Meta Unveils Celebrity Voices for AI Chatbot 👥
Meta Platforms is set to announce exciting audio capabilities for its Meta AI chatbot at the upcoming Connect conference, including celebrity voices from stars like Judi Dench and John Cena. This feature will allow users to choose from five famous personalities, enhancing the interactive experience across Meta's platforms such as Facebook and Instagram.
As the company races to compete with OpenAI and Google in the generative AI space, this move signals a significant push to integrate more engaging elements into digital interactions.
Create custom GPTs that save you time!
Click Image To Play Video 👆
The GPT store is almost here!
To get you prepared, We’re showing you the secrets of creating custom GPTs that will streamline your workflow and save you precious time.
Check out today's AI in 5.
🦾How You Can Leverage:
Apparently we just fast-forwarded 10 years into the AI future.
While us normal humans are all talking about AGI, or Artificial General Intelligence, superhuman Sam Altman just dropped a blog post on super intelligence.
Why’s that a big deal?
After all, y’all wanted it!
// poll results screenshot //
Well, when arguably the most important person in AI says that we’re a ‘few thousand days’ away from superintelligence, we gotta pay attention.
Because it's not every day the CEO of a trillion-dollar-backed AI company predicts the end of human intelligence supremacy.
But that's exactly what happened when Sam Altman dropped his "Intelligence Age" manifesto.
In a few thousand days - that's 8 to 13 years for us mere mortals if we’re going middle of the road- we might be living in a world where AI doesn't just beat us at chess, but runs circles around our entire species.
Is this the dawn of a new era, or just a $6.5 billion fundraising pitch with impeccable timing?
Let's dive into the silicon-laced tea leaves:
1 – Let’s demystify super intelligence 😶🌫️
You didn’t come here to learn?
Then you’re in the wrong place shorty!
Kidding aside, we gotta start elementary AI to truly understand what Sam Altman is talking about when referencing super intelligence. Here’s the most oversimplified version:
AI: The OG. Traditional machine learning and deep learning AI has been around since the '50s. It's your credit card approvals and Netflix recommendations.
GenAI: The new kid on the block. Generative AI has burst onto the scene thanks to powerful Large Language Models like ChatGPT and friends. Fresh faced and popularized in 2022. It's AI that creates stuff - text, images, code, you name it.
ANI: Artificial Narrow Intelligence. AI that's better than humans at one specific thing. Like IBM Deep Blue defeating Garry Kasparov in chess in 1997 or IBM Watson winning Jeopardy! in 2011.
AGI: The holy grail. The ever-moving target. This is an AI that's as smart as humans across the board. We're not there yet, but we're closer than you think.
ASI: The final boss. AI that's smarter than all humans combined in every way. This is what Altman's talking about, and it's either utopia or Skynet.
Try this:
Alright, now that you’ve got the definitions and a loose definition of what’s what, go give Sam’s piece a first read.
Or a second read.
2 – Intelligence is a moving target 🎯
Those definitions above?
They’re constantly changing.
Or, at least how we’ve defined when they’re “achieved.”
Yes.
But, we’ve seen AI advancements go berserk over the past 20 months. The fact that two years ago we didn’t even have AI chatbots and now they can basically do 70% of our work is pretty dang impressive.
Yet….. the definitions (see #1 above) are a moving target. So can we put a finite bar on AI passing human intelligence when even human intelligence is fluid?
We’re not trying to get weird here, but guess what happens as more and more smart humans start handing off a majority of their knowledge work to LLMs?
(Our collectively intelligence may…. go down. Zoinks.)
Sure, having an all-powerful, on-demand genius at our beck and call should make us more intelligent. Like, we should be gobbling up this knowledge like a 2:30 PM egg sandwich when you haven’t eaten all day.
But, truth be told shorties, using AI extensively prolly lowers the overall human intelligence bar.
So, let’s add up all the factors. (DMX voice)
If human intelligence is by and large hitting the snooze button and the ever-changing definitions of what constitutes Artificial General Intelligence is shift-shaping, it’s tough to get a gauge of when we cross some arbitrary intelligence threshold.
Whether that’s AGI or superintelligence.
Try this:
That hot take surprise ya?
It shouldn’t.
This may be the first ‘official’ time we’ve tackled the concept of superintelligence. But, we’ve given AGI the deep dive treatment already. (And a lot of the same rules apply.)
3 – Crazy or prophetic? 🤪
It’s #HotTakeTuesday, right?
You didn’t read this far to be rock-a-byed to sleep, did ya?
Here’s the reality.
Like him or loathe him, we talked today about how Sam Altman is actually/technically/for reals one of the most important people in the world.
That’s not hyperbole.
The largest companies in the world legit rely on OpenAI’s technology. And the future of work largely is impacted by what OpenAI builds.
So when he talks about friggin SUPERINTELLIGENCE in a ‘few thousand days’ we’ve all gotta pay attention.
Is that a wild claim?
Yes.
Is he one of the most qualified people in the world to talk about superintelligence?
Yes.
Will his superintelligence manifesto help him with his next multi-billion dollar fundraising round?
Absolutely.
Try this:
Bookmark this post or flag this newsletter. Come back to this in ‘a few thousand days.’
I’m guessing that we’ll be MUCH closer to superintelligence than we are today. Will it be accomplished?
No clue.
But if we HAD to venture to pick one side, we’d say ASI in 8-13 years at our current rate of AI explosion is definitely feasible.
(And yes, it does feel crazy to type that.)
We just hope it’ll be a more utopian superintelligence future (rainbows and curing diseases) vs. a dystopian one (joblessness and robot-on-human crime.)
Here’s what some of our smart audience said on today’s livestream.
What do you think?
Hit us with a reply and let us know!
Reply