Government Turns a Blind Eye on AI

What's up ya'll, this is AIdeations. The go-to newsletter that takes AI and tech news that slaps and turns it into a no-bs, fun email for you each day.

Here's what we've got in store for you today:

📈 BloombergGPT? Financial LLM's are becoming extremely advanced.

🔎 Clearview AI Tramples on Privacy

📰 News From The Front Lines

🎥 Video Of The Day

🛠 Tools Of The Day

BloombergGPT™: A Game-Changer for Financial NLP Tasks and Applications

Bloomberg has decided to grace the financial world with their latest artificial intelligence (AI) brainchild, BloombergGPT™. This large language model (LLM) is just brimming with excitement to tackle those natural language processing (NLP) tasks that keep the finance industry's gears turning. And to think, it only took a decade of Bloomberg dabbling in AI, Machine Learning, and NLP to make this happen!

But let's not forget the real hero of this story: data. The Bloomberg team has been dutifully collecting and hoarding financial language documents for the past forty years. The end result? A beautiful, monstrous dataset of 363 billion tokens – a feast fit for the most ravenous of AI models. Mix in a dash of 345 billion tokens from public datasets, and voilà: a 700-billion-token training corpus.

What's that, you ask? How does this shiny new AI model perform on actual tasks? Well, according to Bloomberg themselves, BloombergGPT is nothing short of a financial NLP miracle. It outperforms existing models on financial tasks by large margins, while still holding its own on general-purpose benchmarks. Let's all take a moment to appreciate the marvel of technology that is BloombergGPT, shall we?

But what does this all mean for Bloomberg's customers? Well, according to the company's Chief Technology Officer, Shawn Edwards, BloombergGPT will tackle "many new types of applications" while delivering "much higher performance" and faster time-to-market. Gideon Mann, Head of Bloomberg's ML Product and Research team, shared similar enthusiasm, emphasizing the importance of Bloomberg's carefully curated dataset in creating an LLM best suited for financial use cases.

So, without further ado, here are your key takeaways:

• Bloomberg unveiled BloombergGPT™, their AI model specifically trained for the financial domain.

• The model was trained on a colossal dataset of over 700 billion tokens, combining 40 years of Bloomberg's financial data with public datasets.

• BloombergGPT outperforms similar models on financial tasks and competes well on general-purpose benchmarks.

• This model is expected to improve existing financial NLP tasks and unlock new applications for Bloomberg's customers.

• The full research paper detailing BloombergGPT can be found on arXiv: https://arxiv.org/abs/2303.17564.

Government Turns a Blind Eye as Clearview AI Tramples on Privacy Rights

Clearview, the facial recognition firm with a seemingly insatiable appetite for our faces, has carried out nearly a million searches for the US police. Bravo, CEO Hoan Ton-That, for revealing that you've hoarded a whopping 30 billion images, scraped from platforms like Facebook, all without user consent. Because who needs privacy, right?

Let's not forget to give credit where it's due. Clearview has managed to collect fines worth millions of dollars in Europe and Australia for their privacy invasion escapades. Critics are understandably concerned that the police's use of Clearview essentially puts us all in a "perpetual police line-up." But hey, who wouldn't want to feel like a suspect 24/7?

It's astonishing that our government has allowed this level of privacy abuse to persist. Miami Police, for example, proudly admitted to using Clearview's software for crimes of all shapes and sizes, from murder to shoplifting. Assistant Chief of Police Armando Aguilar claims his team uses the system around 450 times a year to help solve crimes. But don't worry, it's not like there are virtually no laws around the use of facial recognition by police or anything... oh wait.

Sure, Clearview's system has proven useful in some cases, like Andrew Conlyn's, who had charges against him dropped after Clearview found a crucial witness. But let's not overlook the potential for mistaken identity, privacy violations, or the fact that Clearview's CEO would rather not testify about the accuracy of the algorithm in court. Because who needs accountability?

So, let's raise a toast to the following takeaways:

• Clearview AI's million searches for US police and their blatant disregard for user consent.

• A jaw-dropping 30 billion images collected without permission, racking up fines for privacy breaches.

• The US government's baffling tolerance for privacy abuse in the name of crime-solving.

• Clearview's occasional success stories, overshadowed by concerns over privacy and algorithm accuracy.

• The glaring lack of laws and transparency around facial recognition use by police, putting our civil liberties at risk.

SOURCE:

📰 News From The Front Lines: 📰

📼 Video Of The Day 📼

🛠️ Tools Of The Day 🛠️

Leiapix - turn any of your favorite images into stunning Depth Animations

ZeroTax -uses advanced artificial intelligence and a team of tax experts to give fast, accurate answers to any tax-related question

Managr - AI sales assistant powered by ChatGPT

CourseAI - Create online courses in seconds

GetFloorPlan - creates 2D, 3D floor plans and 360° virtual tours at no time

Thanks for tuning in to our daily newsletter. We hope you found our tips and strategies for AI tools helpful.

Your referrals mean the world to us. See you tomorrow!

Interested in Advertising on AIdeations?

 Fill out this survey and we will get back to you soon.

DISCLAIMER: None of this is financial advice. This newsletter is strictly educational and is not investment advice or a solicitation to buy or sell any assets or to make any financial decisions. Please be careful and do your own research.