The Future of B2B Sales is Here: DLabs.AI's AI-Powered Sales Specialist

Hello, fellow AI aficionados! Welcome back to our monthly rendezvous. If you've been wondering what the tech titans have been up to in the AI world, you're in the right place.

Let me tell you, it's been a wild ride this month. Just when we thought things couldn't get any more futuristic, these big shots go ahead and pull another AI rabbit out of their hats! But you need to hold onto your hats, folks.?

Because today, I'm not just bringing you the big headlines. I've got an inside scoop from our very own playground at Dlabs.AI. Picture this: an AI-powered Sales Specialist who doesn't need coffee breaks and actually gets the job done. Yes, you read that right. And the best part? You'll be able to hire this tech whiz in-house very soon.

Curiosity piqued yet? Let's dive in ??

Revolutionizing Sales: How We Created an AI-Powered B2B Sales Specialist

With the emergence of ChatGPT, folks who were once comfortably lounging in their jobs are now getting a bit twitchy. Guess what's making them lose their cool? It's the looming question – will AI steal their jobs? And to add a dash of spice, there are even websites popping up where you can calculate the odds of AI elbowing you out of your profession.

Just when you think you've seen it all, AI swoops in with another surprise. If you've been keeping up with my newsletter, you might remember recent news about AI stepping into a CEO’s shoes. I must confess, with the rate at which AI is advancing, it's getting harder and harder to dismiss the idea that it might shake up our professional landscape. I mean, we're already seeing entirely new roles like Prompt Engineer popping up.

However, I've got some seriously cool stuff to share with you. Over at DLabs.AI, we've been tinkering with something that, frankly, feels like it's from the future. You see, we built a tool for ourselves that's been a game-changer for our prospecting tasks. We took ChatGPT, fed it with a bunch of our data, like info about us, our buyer persona, the problems we solve, and the solutions we provide, and let it work. The result? A fully automated machine that spits out laser-focused message sequences for potential new clients, creating personalized messages, follow-ups, and more.

And let me tell you, it's been a revelation. Not only is it doing a bang-up job, but it's also freeing up a ton of our time, which we're putting to good use by crafting better offers for our clients.

But then, we started thinking. What if we could take this to the next level? What if we could build an entire AI-powered B2B sales agent that could handle everything from Client Relationship Management and Sales Performance Tracking to Industry Trends Analysis and Customer Retention?

So we did some digging, checked out some tools, analyzed their AI Autonomy Levels, and guess what? According to our calculation, our so-called AI-Powered B2B Sales Specialist can achieve an autonomy level of 78.3%.

I'll be sharing more details in the coming days, but here's a little sneak peek for you: our prospecting tool will be hitting the market real soon. Moreover, you’ll also be able to hire our AI-powered sales specialist. Got your attention? Good.

Got your attention? If you're interested in learning how we built that, comment 'build' and I will craft a guide on how we made it.

AI Regulation Rollercoaster: OpenAI's Twisty-Turny EU Journey

So, remember how we were talking about AI regulation? Well, Sam Altman, the big guy at OpenAI, recently got everyone's attention with his comments on the European Union's proposed AI rules.?

He was pretty worried that these new regulations might be a bit much, especially for apps like ChatGPT, which could end up under the strictest rules . The funny thing is, he'd been calling for AI regulation over in the U.S. just a while back, so it was a bit of a switcheroo.

While speaking in London, he shared, "We're going to try our best to follow these rules, but if we can't, we might have to cease operations in Europe" . He appeared particularly worried that some aspects of the AI Act's safety and transparency requirements might be technically unfeasible for OpenAI to comply with (such as disclosing which copyrighted material had been used to train their AI systems ).

But here's where it takes a twist. After his comments made waves, Altman did an about-face and announced that OpenAI has no plans to leave Europe after all! He broadcasted this on Twitter, expressing their excitement to continue operating there .

So, yes, it's been quite the rollercoaster with OpenAI and the EU. It will be fascinating to see how this AI regulation saga unfolds, don't you think?

Sources: Forbes, BBC

Unleashing Google's Latest Innovations: Bard, SGE, Project Starline, and PaLM 2

Well, while OpenAI has been actively engaged in AI regulations, Google has recently made significant strides by implementing a range of intriguing advancements. Let's delve into the details and explore some notable highlights:

A Smarter Bard

Google announced that it’s expanding Bard's reach and capabilities: now it's available in over 180 countries, supports English, Japanese, and Korean, with plans to support 40 more languages, and it's equipped with new features like multimodal queries that work with images.?

In addition, Google is integrating Bard with its own apps like Maps, Sheets, Gmail, and Docs, and third-party services like Adobe's Firefly and, soon, Wolfram Alpha. Furthermore, Bard's coding capabilities are enhanced, it can generate, explain, and debug code, and export Python code to Google’s Colab and Replit.?

Source: Google??

New Search Generative Experience (SGE)?

Google is introducing a new search feature, the Search Generative Experience (SGE), which uses artificial intelligence to generate comprehensive answers to your queries. This AI-driven feature generates comprehensive, full-sentence responses, making your online search feel like an interactive conversation.?

With SGE, finding complex answers is simplified, effectively streamlining your search process. However, this innovation doesn't come without its challenges. Ensuring the AI doesn't merely summarize content from other sites, managing to integrate ads into AI answers, and minimizing bias and misinformation are all aspects that need careful handling.?

To ensure transparency, SGE pairs its AI-generated answers with source links, though this has raised concerns about potentially limiting exposure to original sites. I'm intrigued by the potential impact of this shift on marketing and website performance.

If you're curious to see this in action, I highly recommend a video that made quite an impression on me:

Source: Google?

Project Starline?

Have you ever wished you could teleport to see your loved ones? Well, Google's Project Starline is making that wish come true, sort of. It's not actual teleportation, but it's pretty darn close. Picture this: you're sitting in a booth, and boom, there's a 3D, life-sized image of your best friend sitting across from you.?

The level of immersion is incredible, all thanks to a mix of sophisticated tech like computer vision, machine learning, spatial audio, and real-time compression. The future of digital interaction is here, and I can't wait to see how this will transform the way we communicate.

Source: Google

PaLM 2

Last but not least, let's talk about the brain of the operation, PaLM 2, Google's new language model. This whizz of an AI is all set to take on systems like OpenAI’s GPT-4.

"PaLM 2 models are stronger in logic and reasoning, thanks to broad training in logic and reasoning," said Google CEO Sundar Pichai. "It’s also trained on multilingual texts spanning over 100 languages."

The system is available in four sizes, named Gecko, Otter, Bison, and Unicorn, from smallest to largest, and has been fine-tuned on domain-specific data to perform certain tasks for enterprise customers. The lightest version, Gecko, can run on mobile phones (because who doesn't want a pocket-sized, super-intelligent AI, right?).

Within Google’s own domain, PaLM 2 is already being used to power 25 features and products, including Bard, the company’s experimental chatbot. Updates available through Bard include improved coding capabilities and greater language support.?

It’s also being used to power features in Google Workspace apps like Docs, Slides, and Sheets.

Alright, let's get some comments from the wizard behind the AI magic at Dlabs.AI, Filip Skurniak .?

Here's what he had to say:

It’s pretty cool seeing powerful language models like PaLM 2 showing up on our smartphones. This trend is a big deal for a couple of reasons. First, it’s a win for privacy: our data stays on our device. Second, we can use these features offline, so no Wi-Fi, no problem.

What’s more, if mobile LLMs keep improving, we might see a lot more apps switch to a conversational interface. Imagine how that could change our everyday app experience!

By the way, with all this going on, what’s the scoop on Google Assistant? Any new updates or changes there?

Source: Google??

—-

So there you have it! Google has been quite the busy bee, with its chatbot learning new languages, its search turning into a chatterbox, and its new AI model flexing its skills. It's all just a month in the life at Google!

Meta Harnesses the Power of the Bible: AI Models That Recognize 1,000+ Languages

Let's switch gears and catch up with Meta, which has been flexing its AI muscles in the realm of language. Meta's AI is now a polyglot, with speech recognition and generation capabilities in over 1,000 languages.?

As Meta says, that's a big step towards preserving the linguistic diversity of our world. This move could be a game-changer, enabling developers to create speech applications that can communicate with users in their native tongues. Keep in mind, our world has roughly 7,000 languages, but current speech models cover only about 100.?

The main challenge? Limited labeled training data for the majority of languages. Meta tackled this head-on, retraining an AI model to learn speech patterns directly from sound, bypassing the need for extensive transcriptions. For the training, they used two unique datasets: one with audio and text of the New Testament in 1,107 languages and another with unlabeled New Testament audio in 3,809 languages.

The result? Models that can converse in over 1,000 languages and recognize more than 4,000! However, it's not all wine and roses; these models come with a sprinkle of error and potential bias.?

Plus, using religious texts for training? Well, I must say, it's a bit controversial, isn't it?

Source: Meta AI

AI Dives into Biology: Unearthing Rare DNA Sequences and Revolutionizing Genomic Research

AI is reshaping the world, and it's not just about business and productivity tools. It's delving into the deepest secrets of our existence – our very own genes. Spearheaded by Professor James T. Kadonaga from the University of California San Diego, a remarkable team has brought the power of machine learning, a branch of AI, into the realm of biology.?

Their mission? To unearth the rare DNA sequences that control the activation of our genes. This cutting-edge exploration, led by the School of Biological Sciences Professor James T. Kadonaga, has identified the downstream core promoter region (DPR), a DNA activation code that plays a role in approximately one-third of our genes' operations.

Building on this breakthrough, Kadonaga and his team have leveraged machine learning to pinpoint "synthetic extreme" DNA sequences with specialized functions in gene activation. This process involved comparing the DPR gene activation element in humans and fruit flies, leading to the identification of rare DPR sequences that are active in humans but not in fruit flies, and vice versa. This could have far-reaching implications for biotechnology and medicine, allowing for the identification of synthetic DNA sequences with potential practical uses.

Just as AI technologies like ChatGPT are revolutionizing our daily lives, they're also reshaping scientific research in ways we couldn't have dreamed of before. And this is just the beginning. As Kadonaga notes, "Biologists are probably at the very beginning of tapping into the power of AI technology."

Source: Today

First Person Arrested for Misusing GPT Technology

Well, in the end, it might not be groundbreaking news, but certainly, something that made me rub my eyes in astonishment….

While the EU and the US are contemplating legal regulations concerning AI, China has made headlines by arresting an individual for utilizing GPT-based technology. The Chinese police apprehended Hong for allegedly fabricating a news story about a train crash using ChatGPT.?

This incident marks the first enforcement action taken under China's novel law about artificial intelligence. And the arrest highlights the Chinese government's determination to regulate and control the usage of AI as technology advances.

It's worth noting that this arrest was made possible by China's introduction of the "deep synthesis technologies" law earlier this year. This law encompasses AI-generated text, images, video, and other media, explicitly prohibiting the dissemination of fake news through AI.

Oh, but isn't it truly remarkable how their government unfailingly ensures the publication of truth and maintains utmost reliability by introducing rigorous censorship on the internet? Well, censorship surely guarantees the authenticity and trustworthiness of information…

Source: CNBC

______

That concludes today's newsletter.?

I hope you enjoyed the read. If there’s anything you'd like to discuss, feel free to share your thoughts in the comments.

See you in June!