What’s Up at the AI and No-Code Frontier?

I’m Pavel, the CEO of a no-code company called Directual. Here, I’ll share my view on the latest on the AI frontier, as seen by a no-code platform that is integrated very tightly with OpenAI.

The whole AI shebang isn’t new. It’s just that after OpenAI released ChatGPT and Dall-E things really started to spiral out of control. Now, you get some groundbreaking AI news every other day at the very least.

Since Directual is a no-code platform tightly integrated with OpenAI, we also observe news headlines with bated breath. After all, who knows what will happen next? Is it the end of no-code platforms, as it could be for a whole gazillion businesses and jobs out there? Or is it the other way around? Anyway, let’s see what’s happening so far instead.

What’s up on the AI frontier?

Plenty!

Investments in AI so far

The world of generative AI is heating up faster than yet another Kardashian feud. High costs, ambitious goals, and external pressures have set the stage for a wild investment spree. It's like everyone's competing in the Olympics, except instead of medals, they're aiming for that sweet, sweet funding.

All this is happening during times when venture capital isn’t exactly thriving. Interest rate hikes by the Federal Reserve have sent tech valuations into a nosedive. To make matters worse, tech companies have been laying off workers left and right, with over 185,000+ jobs axed in 2023 alone.

Investors are flocking to the generative AI sector like moths to a flame, however, drawn in by its immense potential. And boy, have there been some massive funding rounds lately:

OpenAI scored a ridiculous $10 billion from Microsoft in January. In fact, OpenAI's funding alone exceeded the entire investment in generative AI companies for 2022, according to CB Insights. Talk about making it rain!

• Jasper bagged $125 million in October and is now worth $1.5 billion. Not too shabby.
• Character.ai collected $150 million.
• Stability AI snagged $101 million and hit a $1 billion valuation.

No wonder, really. Building a model like that is EXPENSIVE.

According to Mike Volpi, a partner at Index Ventures, you're gonna need at least $500 million to develop a large language model like the one powering ChatGPT. A great lot of it is spent on computing power. That $10 billion investment into OpenAI by Microsoft? About half of it will go into Microsoft Azure.

It’s also a big reason why the number of players in the AI market are likely to remain low—good luck getting into this kind of market unless you have a fat stack of cash laying around doing absolutely nothing else.

So, if you're feeling FOMO, don't worry, there's still time to get in on the action in terms of investments into the frontrunners. Just like the tech sector during COVID-19, it seems that AI companies are set to be prize winners for many denizens of Robinhood and the like.

Regulations rolling in

Thanks for this totally-not-doomsaying post by OpenAI’s CEO, if someone wasn’t scared of what AGI will do, they certainly are now. For legislators around the globe, AI is also about to become a tremendous headache. How do you secure the job market? How do you ensure private data stays private? So many different avenues to tackle, and so little time.

The European Union is stirring up the pot with new regulations on AI development (how about some irony: let ChatGPT give you a summary of this act. Wouldn’t that be a laugh?) and deployment, and it looks like OpenAI and other companies may have to cough up information about their use of copyrighted materials in training their AI. The EU's AI Act will also classify different AI technologies according to the risks they pose, and some chatbots could be slapped with a "high-risk" label. Figures, really.

Recent debates over chatbot regulations suggest that the AI Act won't be finalized until at least 2024, but EU member states are already making their own policies in the meantime. This means companies like OpenAI will need to be extra careful as they navigate a tangled web of governance criteria.

That may be a reasonable explanation for why Reddit users are complaining that ChatGPT is getting “nerfed” with every single update (although you can still try to gaslight it into telling you what you want).

Hilarious! Akhem, anyway…

Lawmakers are also proposing that chatbots like OpenAI's ChatGPT should disclose to users that they're not human, while also requiring AI platforms to provide greater transparency on the training of their models. In short, EU lawmakers are cracking down on the rise of generative AI technology, which is great news for people but not so much for companies like OpenAI.

In fact, Italy banned (and then unbanned) ChatGPT at the beginning of April. OpenAI was given 20 days to sort this out, and they did. They announced that its banned chatbot is finally allowed to come back online, thanks to a few privacy-related changes made.

The changes include a new article that explains how the bot collects and uses data, an opt-out form for EU users, a more visible privacy policy, and a tool to verify users' ages in Italy.

A call for a standstill

Slightly old news, but nonetheless. 

Over a thousand people, including Elon Musk and some big-shot researchers, have signed an open letter warning that the risks of AI are imminent if we don't slow down. They're calling for a six-month pause in all "giant AI experiments".

In other words, they want to put a hold on developing powerful AI systems until we're sure they won't ruin everything. Because let's face it, we've hit pause on other potentially catastrophic technologies before (ozone layer destruction saga, anyone?). And right now, it seems like we're in an "out-of-control race" to create digital minds that we can't even control. It’s so bad, right?...right?

It is rumored, however, that all of these big people signing the open letter and pushing the agenda to pause AI development are doing it to stall OpenAI’s progress, which is vastly outpacing everyone else at the moment. Because…my oh my, Elon has his own AIin the works! 

Time will only tell if everyone was right, but 2023 is bound to be a wild year for sure.

NVidia’s text-to-video…wow.

The nutty professors over at Nvidia created a Latent Diffusion Model that can make high-quality videos up to 4.7 seconds long, all from text. What’s more, they went ahead and used open source Stable Diffusion model and DreamBooth to tune it up even more! This just proves that we're barely scratching the surface of what's possible on this front.
I mean…just look at it.

AI music setting the whole industry on fire..figuratively and literally

Universal Music Group (UMG) is playing whack-a-mole with AI-generated songs that imitate the voices of Drake and The Weeknd. The company filed a copyright infringement complaint resulting in the removal of "Heart on My Sleeve," a viral AI-generated song from streaming platforms, which gained over 15 million streams. Drake himself slammed the track on Instagram, calling it "the final straw," but legal experts are questioning UMG's claim since generative AI isn't explicitly addressed in copyright law. The original used to be here, RIP.

Here’s a version that is still alive. 

It kinda slaps!

While UMG is trying to take the moral high ground, it appears to have selective outrage as it remains mum on other AI-generated songs that feature its artists, such as a cover version of Ice Spice's "Munch (Feelin' U)" and Beyoncé's "Cuff It" remix.

The end of the job market is nigh

Seems like executives are not sure what to make of generative AI just yet. 

On one hand, 65% of them believe it will have a big impact on their companies, but on the other hand, nearly the same percentage thinks it's still too early for that impact to happen.

Almost 4 in 10 of these executives are afraid that generative AI could turn their employees into social outcasts, decreasing social interactions and human connections.

And while some are still on the fence, corporations like PwC are not wasting any time, investing a whopping $1 billion in AI over the next 3 years. Looks like they're taking the "go big or go home" approach when it comes to AI investments.

The layoffs have already begun, and the most common topic on the OpenAI subreddit is “Are we going to see a shortage of jobs or not”. Or the likes of it. Opinions vary since some say it’s the biggest thing since the industrial revolution, and others don’t believe anything drastic will happen. I mean, look at the government systems. Germany still sends faxes. Computers are still on Windows XP. We’ll be fine, for the most part, yeah?

What does it mean for no-code platforms?

So, will no-code platforms like Directual be out of a job because of the rise of AI? In short, hell no. No-code platforms are only getting stronger, especially when they integrate with AI. Don't believe me? Let's take a closer look.

First off, let's talk about the learning curve. 

Even the savviest programmers may struggle to interact properly with AI, and getting it to do what you want is like playing a game of 20 questions. With no-code platforms, however, you can just stack pre-made blocks together like Lego and your app will magically work.

Coding with AI also means you'll be constantly copying and pasting snippets of code, testing them out, and praying they work. Not to mention, you'll need a lot of documentation to make sure everything's running smoothly. With no-code platforms, everything's already laid out for you in a single workflow, and templates can help you get started in just ten minutes.

Just like that. I swear I didn’t write this entire article just to arrive at this video.

What happens when you need to scale, though? That’s the big thing.

AI has its limits in terms of memory and response length (even the likes of GPT4), which means you might end up with a headache trying to break up your problem into smaller chunks. Plus, if you need to copy logic from one scenario to another, good luck. AI will likely make errors or drift off into space. With no-code platforms, you can reliably stack everything together and scale your app without hitting any roadblocks.

Thus, no-code platforms will be just fine, for one reason: you should be using both them and AI. No-code platforms can help you create apps faster and refine your processes, while AI can give your app advanced chat-based functionality and help with more complex tasks. Just make sure to say “thank you” to your AI overlord after you’re done with your query. No such requirements from us. 

Afterword

The world of AI will keep on accelerating faster. Who knows what we’ll see tomorrow: a utopia where no one needs to work (machines go in, men go out!), or will it be Skynet eradicating humans like the little noxious pests that they are? 

For the time being, one thing is certain: you’ll be wise to invest into AI, and keep a close eye on what’s happening with it. Future-proof your job, your projects, and your company, because the competition’s landscape may look very different, very quickly.

Cheerio.