🤷 So Have we achieved AGI? The OpenAI Ethical team thinks so.
Plus: Inflection dropped one of the Largest LLMs ever, AI Influencers are making money, AI Tolls to meditate and relax, and more.
Hope you had a Great Thursday 👍,
We are back with a short letter today. What is Project Q*? Have we achieved AGI? We share about AGI and the relationship between that and OpenAI drama. Also, Did you eat and drink a lot yesterday? had crazy conversations over Thanksgiving? we have the AI tools for you to relax and meditate on it. AI Influencers are here to stay and getting paid.
This is what’s going on:
So Have we achieved AGI? and What is Project Q* by OpenAI?
Inflection dropped one of the Largest LLMs ever.
AI Influencers are Making Bank without the Headaches.
Cheat Sheet of the Week
🤔 What Does that even Mean? (Deep Learning)
🧰 AI Tools of the Day (Relax / Meditate)
💰 Follow the Money
"Innovation is the outcome of a habit, not a random act." – Sukant Ratnakar
💪🏽 Important AI News and Trends.
AI is not Human, please remember (Vox)
Inside OpenAI, the Architect of ChatGPT, featuring Mira Murati (YT)
A.I. Belongs to the Capitalists Now (NYT)
OpenAI researchers warned the board of AI breakthrough (Reuters)
ChatGPT's voice chat feature is rolling out to free users (Engadget)
🌐 Web3 and Other Tech news.
Tesla ignored Autopilot’s fatal flaws, judge says evidence shows (Ars)
Tech getting better at smelling (BBC)
Binance users have withdrawn close to $2 billion in the last 48 hours (CNBC)
Spotify will no longer compensate artists/songs with less than 1k streams (HB)
Tesla introduces new fee to limit congestion at Superchargers (TheVerge)
So Have we achieved AGI? and What is Project Q* by OpenAI?
The Quest for AGI Amidst Ethical Concerns.
OpenAI has been thrust into the spotlight with its controversial Project Q*, a secretive initiative aimed at developing artificial general intelligence (AGI). This ambitious project promises AI capabilities that could surpass human intelligence, but it has also sparked serious ethical debates within the company. Researchers have expressed concerns about the potential risks such advanced AI could pose to humanity, leading to the temporary removal of CEO Sam Altman from his post.
Q* has demonstrated its potential by outperforming humans in basic math, suggesting advanced reasoning abilities. Despite the concerns, Altman's leadership has been reinstated, indicating OpenAI's commitment to forging ahead with AGI development. This advancement is seen as a significant technological leap, yet it raises important questions about the societal impact and the future of employment. As OpenAI navigates this complex landscape, the balance between innovation and ethical responsibility remains a critical challenge. We will continue exploring this intersection and help our readers be ready for anything that comes ahead.
Inflection dropped one of the Largest LLMs ever.
Inflection has announced the completion of Inflection-2, an AI model that now ranks as the second most capable large language model (LLM) globally and leads in its compute class. Allegedly, Inflection AI LLMs have been trained only using publicly available data and proprietary data.
This model surpasses its predecessor, Inflection-1, in knowledge accuracy, stylistic versatility, and reasoning ability. Trained on 5,000 NVIDIA H100 GPUs, Inflection-2 outperforms Google's PaLM 2 Large and others on several AI benchmarks. It promises more efficient performance with reduced cost and increased speed, thanks to improved GPUs and optimized inference processes. The company emphasizes its commitment to safety and trustworthiness, aligning with global governance efforts. Before its integration into the Pi platform, which I have been a fan of and use on my mobile device regularly, Inflection-2 will undergo alignment to ensure it is a beneficial and secure personal AI. This model represents a significant step toward Inflection’s goal of providing a personal AI for everyone.
AI Influencers are Making Bank without the Headaches.
Meet Aitana López an AI-generated Spanish model earning up to €10,000 monthly, with an average of €3,000. Created by Rubén Cruz of The Clueless agency during a downturn in client work, Aitana offers a reliable and controlled alternative to human influencers. Despite being virtual, she's gained over 121,000 Instagram followers and has even received date requests from celebrities. Her creators curate her digital life, reflecting trends to appeal to followers. While AI models like Aitana may reduce advertising costs and democratize branding, they raise concerns about perpetuating unrealistic beauty standards and sexualization in media.
Cheat Sheet of the Week
Popular AI Apps of 2023 - Source
💰Follow the Money
$6.3M, raised by Handraise, Austin Austin-based AI-powered PR platform.
$4.6M in Seed funding by CreateSafe, an LA-based AI music tech studio.
$3 Million, funding for Birdeye, AI Marketing Platform.
$1.8 Million by Layer AI for non-gaming work such as marketing and advertising.
🤔 What Does that even Mean?
Deep Learning:
A subset of ML that uses neural networks with three or more layers. These neural networks attempt to simulate the behavior of the human brain—albeit far from matching its ability—allowing it to “learn” from large amounts of data.
Example: Consider an app that identifies flowers from photos. You train it with many labeled flower images (Basically what we do with CAPTCHAS). The app's deep learning system learns details like petal shape and color. Once trained, it can accurately name flowers it's never seen before by recognizing these details.
🧰 AI Tools of the Day
Relax / Meditate.
Vital - Enter what you want help with and get a guided mindset session.
Guided - Prompt it and generate a guided meditation tailored to your needs.
Pneuma - Meditation, breathing, and focus app.
Clearmind - Therapy platform to measure and elevate your emotional health
Download over 250+ Tools free here.
Another great post. I'm happy about the Inflexion news - have Pi installed, but rarely use it because it seems limited. Maybe this will change my view.