- The AI Plug
- Posts
- #20 AI decoding our thoughts
Happy Monday, people.
We’re closing in on 5,000 subscribers, and would love to thank YOU for being a dedicated early reader of The AI Plug - we wouldn’t be here without you.
Some exciting figures and concepts in this issue, from billion-dollar investments to potential mind-reading technology. So without further ado;
In today’s newsletter:
Top News: AI mind-decoding, Meta abandons the Metaverse and Microsoft’s investment qualms
The Rundown: AI scammers destroying businesses, Anthropic’s $5B plan, and AI fund outperforms rest of the market
Tips, Tricks & Tools: Voice-powered app building, business idea feedback, and realistic interview practice
What’s Your Opinion?:
One’Sloss is another’s gain: VPN providers to profit in face of EU GPT bans
Top News 🔝
Three biggest stories if you’re in a rush
Ranking within the top 1% for engagement among 23 million research outputs, a recent research paper from Osaka University details how AI successfully decoded subjects’ brain activity, translating it into high-fidelity images of what the subject was seeing. Using Stable Diffusion (SD), researchers Takagi and Nishimoto built the model to translate a person’s brain activity into a visual format.
“This is not mind-reading…we can’t decode imaginations or dreams…but of course, there is potential in the future.”
Presented images (top row) and AI-reconstructed images (bottom row) from subjects’ brain activity. REPLY TO THIS EMAIL WITH “AI IMAGES”, AND WE’LL SEND YOU THE FULL PAPER.
Takagi and his team have started working on a much more advanced image-reconstructing technique, one which can be applied to novel subjects and can recognize images with which the model is unfamiliar.
- - - - - - - - - - - - - -
Mark Zuckerberg’s milestone October 2021 rebranding of Facebook to “Meta” seemed to indicate the company’s future strategy - investing billions into the metaverse, to create a “highly immersive” and “fully embodied” internet.
But as AI becomes an ever-more lucrative money-printing industry, Meta is dropping the metaverse like a hot potato, instead spending their time developing new generative AI teams and heavily investing in AI advertising.
We can most certainly expect exposure to hyper-targeted, highly personalized, AI-generated ads in the near future, but as for Meta’s metaverse?
Perhaps there’ll be a re-rebrand. Arty?
- - - - - - - - - - - - - -
OpenAI is one of the hottest private companies in the world, after the San Francison-based startup introduced ChatGPT and received a cumulated $13B investment from Microsoft.
Part of this investment agreement was that Microsoft is the exclusive provider of computing power for OpenAI’s products and research. As thousands of new startups and global companies scramble to integrate ChatGPT and other OpenAI products, Microsoft’s cloud servers could generate over $30B in new annual revenue.
But despite the early (impressive) success of OpenAI, it’s early days. Competitors to OpenAI are securing billions in investment, including Anthropic, and Meta and Google will not remain quiet. Musk has expressed his disdain for OpenAI’s profit-generating pivot and has made clear intentions to develop his own competitor.
What remains to be seen is how much Microsoft can capitalize on their and OpenAI’s headstart to secure their massive investment safely.
The Rundown 🐂
Deeper dive into trending stories
AI scammers can and are destroying businesses, stealing over $11m from companies and individuals with voice-impersonation scams. (link)
Global X Robotics & AI ETF, the world’s largest AI fund, is up 23% so far in 2023, outperforming the Nasdaq 100. (link)
Six experts share how they ise next-gen AI to make their work easier, more fun, and more profitable. (link)
Anthropic plans to build self-teaching AI in a $5b, 4year plan to overtake OpenAI. (link)
Samsung employees ask GPT to generate notes from a private meeting recording, leaking sensitive company information to the chatbot. (link)
Luma AI is releasing its video to 3D model API - something notoriously difficult to achieve. (link)
Prediction: “If somebody builds a too-powerful AI, under present conditions, I expect that every single member of the human species and all biological life on Earth dies shortly thereafter”. (link)
Rabbi Google could be changing the very face of religion. (link)
Endometriosis diagnosis without the need for surgery is possible with AI within the next two years. (link)
Top three AI stocks to buy in April. (link)
Bank of America report indicates AI could boost the world economy by over $15T within the next six years. (link)
AI + DNA are unlocking the mysteries of the global supply chain, showing consumers and companies where products really come from. (link)
Tips, Tricks & Tools ⚒️
Discounts, trials and productivity-boosting hacks
Codewand - Build and deploy brand new web applications from scratch, using just your voice. (link)
Adobe Firefly - Make an infinite range of unique image creations. (link)
ChatBotKit - The fastest and more simple way to design and launch an advanced chatbot. (link)
Superhuman - Rank and sort important emails according to your behavioural patterns. (link)
Venturus - Get instant critique and feedback on any business idea. (link)
YCInterview - Choose an interviewer and practice trial interviews for free! (link)
One’s loss is another’s gain
Someone always wins
Following Italy’s lead, Germany, France, and other EU countries are considering a ban on ChatGPT, citing privacy concerns.
As nations move toward banning AI tools, citizens move towards finding ways to circumvent such bans.
In this case, it’ll be by using VPNs.
Now’s a great time to own a VPN service.
What’s Your Opinion?
Tell us what you think
In light of more EU nations discussing bans on ChatGPT, what’re your thoughts?
Do you think bans are the right way for governments to regulate AI?
Thank you to readers Teeabs and JessyM for your responses to “Should we talk about AI tragedies.” Your insight helps us mould this newsletter going forward!~
Spread The Word
Share and earn epic rewards
Lulu’s gonna help us - so why don’t you?
Tell one person about our new newsletter design by sending them the link below ⏬.
If you do, we'll send you a totally free, extensive course on prompt engineering, where you will learn to:
Get to most out of GPT and Midjourney
Automate workflows
Free up your time
Do more cool sh*t instead of working