Welcome to Verses Over Variables, a newsletter exploring the world of artificial intelligence (AI) and its influence on our society, culture, and our perception of reality.

Tool Update

Claude 3.5 Sonnet by Anthropic with Artifacts: New Release

What it is: On Thursday, Anthropic dropped its latest bombshell, Claude 3.5 Sonnet. This isn’t your grandmother’s chatbot. It's a lightning-fast, multilingual math whiz that can code circles around its predecessors and rivals. And the kicker? It's cheaper than its beefier sibling, Claude 3 Opus. Talk about a bargain-bin brainiac.

But wait, there's more! Sonnet comes with a party trick that will make even the most jaded tech bros sit up and take notice. Enter "Artifacts," the feature that turns your AI interactions into a live-action spectacle. Picture this: You're chatting with Claude, and suddenly, BAM! A dedicated window pops up, and you're watching this digital dynamo whip up code snippets, documents, and designs faster than you can say "Silicon Valley." Want a game? A chart? Some mind-bending visualization? Just ask, and watch the magic unfold before your very eyes. It's like having a tiny, hyper-efficient dev team trapped in your browser. The possibilities are endless, limited only by your imagination and, let's be real, your daily message limit. (Here are some examples of the model in action from Anthropic. Here are some examples of use cases from other creators.)

How we use it: We’ve been putting Claude 3.5 Sonnet through its paces: to summarize PDFs with slick visualizations and whip up dashboards from Excel data faster than you can say "pivot table." But the real game-changer? Interactive games for our classes and workshops. We're still figuring out how to integrate this digital wunderkind into our teaching fully, but one thing's for sure – the classroom just got a whole lot more interesting.

Some of Our Favorite Tools

Eleven Labs: Sound and Fury

What it is: ElevenLabs created a variety a AI-powered voice tech that can read, translate, and even impersonate with uncanny realism. It's a vocal revolution that's making waves in everything from audiobook production to political campaigns, promising (or threatening, depending on your view) to reshape how we create and consume audio content. Their flagship offering? A text-to-speech tool that doesn't just read words, but performs them with all the nuance of a seasoned actor. We're talking AI voices that can laugh, pause for dramatic effect, and even nail the pronunciation of "minute" based on context. They also have a voice cloning feature lets you upload a few minutes of audio and - presto! - you've got a digital doppelganger ready to say whatever you type. For the less narcissistic, there's a Voice Library stocked with pre-made voices. The pièce de résistance? A dubbing tool that can translate videos into 29 languages while preserving the original speaker's voice. It's like a United Nations interpreter on steroids. And for the cautious types, ElevenLabs has whipped up an AI Speech Classifier. This digital detective can sniff out whether an audio clip is one of their AI creations - a sort of audio watermark for the deepfake era. Recently, ElevenLabs has been on fire launching new tools including a Text to Sound Effects and a Voiceover Studio.

How we use it: Mostly, we use ElevenLabs for our video content as we hate hearing our own voice. But if you’d like to play around, you can also use their Custom GPT.

We’ll be talking about our favorite tools each week but here is a list of what tools we use most for productivity: ChatGPT 4o (custom GPTs), Midjourney (image creation), Perplexity (for research), Descript (for video, transcripts), Claude (for writing), Adobe (for design), Miro (whiteboarding insights), and Zoom (meeting transcripts, insights, and skip ahead in videos).

Intriguing Stories

AI’s New Safety Net: Ilya Sutskever, the enigmatic co-founder and former chief scientist of OpenAI, has finally lifted the veil on his next act: Safe Superintelligence Inc. (SSI). This isn't just another tech startup; it's Sutskever's audacious bid to create the holy grail of artificial intelligence – a safe, superintelligent system that could make current AI look like a pocket calculator. Teaming up with ex-Apple AI lead Daniel Gross and former OpenAI engineer Daniel Levy, SSI is laser-focused on its moonshot goal, with no plans to dabble in commercial products or services. (It's a throwback to the original ideals of OpenAI, before the allure of ChatGPT's viral success changed the game and its non-profit status.) Sutskever envisions a future where AI safety isn't just a buzzword or an afterthought, but the very foundation of the technology. He's talking about "nuclear safety" level precautions, not just feel-good "trust and safety" measures. While funding details remain under wraps, the company's pedigree suggests no shortage of interested investors. As Sutskever cryptically describes his recent experiences at OpenAI as "very strange," all eyes are on SSI to see if it can deliver on its ambitious promises.

Apple’s AI Hurdles: Apple's ambitious AI rollout faces significant challenges on multiple fronts. In China, Apple is encountering the unique challenge of its AI partner, OpenAI's ChatGPT, being banned. In a move raising eyebrows and probably blood pressures, Apple is cozying up to some unlikely bedfellows: the company is exploring potential arrangements with local tech giants such as Baidu and Alibaba to provide chatbot services for the Chinese market. In Europe, Apple appears to be playing chicken with the regulators, withholding the rollout of its new Apple Intelligence features in the 27 EU countries. Apple cites concerns that its new AI features might violate the Digital Markets Act. Apple won’t risk its systems security or enter the market unless it knows whether it is at risk for a fine or a system change. And even stranger, we’d heard many rumors before the WWDC that Apple was in talks with a variety of tech firms to integrate their AI with Apple’s (including Google’s Gemini), but Apple has also been discussing a potential AI partnership with longtime rival Meta, highlighting the company's efforts to diversify its AI offerings and catch up in the rapidly evolving AI landscape.

AI’s Starving Artists: As AI companies like OpenAI continue their march towards world domination (or at least a fat bottom line), they're facing pushback from the creative community. Authors and artists are crying foul over the unauthorized use of their work to train these silicon-brained behemoths. OpenAI's attempt to appease creators with its upcoming "Media Manager" tool has been met with all the enthusiasm of a participation trophy at the Oscars. This week, OpenAI CTO Mira Murati even said the quiet part out loud: “Some creative jobs will go away, but maybe they shouldn’t have been there in the first place,” while trying to explain AI as a creative, collaborative tool. The rise of AI is hitting some freelancers where it hurts most - their wallets. From copywriters to concept artists, the gig economy is feeling the squeeze as clients turn to artificially intelligent alternatives. And Adobe had to overhaul its terms of service, again, to explicitly state that it won’t train AI on customers’ work, which obviously didn’t quell the backlash.

Digital Déjà Vu: In a twist that would make even the most seasoned plagiarist blush, AI search startup Perplexity has been caught with its digital hand in the cookie jar. After WIRED exposed the company's shenanigans - from surreptitious web scraping to generating pure fiction - Perplexity's chatbot decided to summarize the very article calling it out, copying chunks verbatim in its Pages, emails and podcasts. This billion-dollar darling, backed by Jeff Bezos and Nvidia, now finds itself in a legal gray area that's about as clear as a muddy puddle. While we’d call this copyright infringement, the founders shrug it off as a mere trifle in the Wild West of AI. As the dust settles, one thing's for sure: Perplexity's legal team is about to rack up more billable hours than a Silicon Valley startup during an IPO frenzy.

— Lauren Eve Cantor

thanks for reading!

if someone sent this to you or you haven’t done so yet, please sign up below so you never miss an issue.

if you’d like to chat further about opportunities or interest in AI, please feel free to reply.

if you have any feedback or want to engage with any of the topics discussed in Verses Over Variables, please feel free to reply to this email.

we hope to be flowing into your inbox once a week. stay tuned for more!

banner images created with Midjourney.