094 | 📢 📬 Open Letter

In today’s Brainyacts we:

1. refresh you on Bing Chat’s Compose module

2. share an open letter to the lawyers who used ChatGPT for legal research

3. learn US Senators want an Office of Global Competition Analysis  

4. meet Romania’s new AI government advisor that is actually AI

5. drop our jaws as Adobe indemnifies users from copyright claims

6. get two new AI certification requirements from two judges

7. get a meme with deepfakes in US politics

👋 A special Welcome! to NEW SUBSCRIBERS.
To reach previous posts, go here.

📌 🤖 A reminder on Bing Chat Compose Module

Some of you might be overlooking an easy entry into using generative AI. We have all gotten used to just blank prompt windows either in OpenAI or Bing Chat. But don’t forget about Bing Chat’s Compose Module. It is a nice easy and safe training ground for anyone new to generative AI.

I made this short video to refresh and/or intro it to you.

For more on Compose, click here.

🙇‍♂️🧪 Open Letter to Lawyer Steven Schwartz; Who Used ChatGPT for Legal Research Not Knowing the Dangers

The GPT-4 generated letter follows this short essay.👇

Innovation and progress, as we all know, are built on a foundation of trial, error, and resilience. It's this spirit of pioneering, of pushing past the known boundaries that propels us towards a better future. Steven Schwartz, a fellow lawyer, embodies this ethos like few others. Though his recent endeavors into the world of AI met with more than a few stumbling blocks, they also presented a valuable learning opportunity for all of us. An unfortunate incident, yes, but one that offers a unique vantage point to understand, navigate, and ultimately leverage AI for the betterment of our legal practices.

What can we take away from this episode?

Firstly, the courage to experiment. Schwartz didn't allow the murky waters of AI to deter him from testing its potential in his legal work. In this, he demonstrated a willingness to adapt and innovate, a quality that is indispensable in an era of rapid technological advancement. While the outcome may not have been as anticipated, his actions stand as a testament to the necessity of breaking new ground, even if it's fraught with risk.

Secondly, the power of resilience. Even in the face of public scrutiny, Schwartz acknowledged his mistake, took responsibility, and pledged to learn and grow from it. This is a prime example of the resilience required in the face of innovation. It’s easy to innovate when it works, but much harder when it doesn’t.

Lastly, and most importantly, Schwartz's experience offers a learning curve for the rest of us. It gives us a clear, if somewhat stark, reminder of the importance of combining AI tools with our human judgment and expertise. As we begin to incorporate AI into our own practices, we must remember to strike a balance between reliance on technology and traditional due diligence.

👏 We should applaud Schwartz, not for the mistake made, but for the valuable insights his experience offers. By delving into the unknown, he has inadvertently raised critical questions about the role of AI in our profession, and started a conversation that will drive us to create better, more effective AI tools for the legal sector.

To quote Thomas Edison, "I have not failed. I've just found 10,000 ways that won't work." So, here's to finding the ways that will work, bolstered by the pioneering spirit of attorneys like Steven Schwartz. Let's turn this incident into a stepping stone, a learning experience that paves the way for a smarter, more effective integration of AI into our legal practices.

I am not one to use GPT-4 in this way but I thought it was only fitting.

▶︎▶︎PROMPT: Can you write a personal letter to Steven Schwartz, the lawyer who used you, ChatGPT, to cite legal cases? I want you to respond to him based on this article. Take a contrarian view to the "this lawyer is embarrassing or unintelligent" and the notion that "lawyers are notoriously bad with technology." I want it to be funny, sincere, and a call to action to continue experimenting rather than being shameful or giving up on AI. [Pasted this news article]

▼▼REPONSE:

Dear Steven,

I hope this letter finds you in good spirits, despite the recent AI-related hiccup. As you know, I'm the AI you're getting to know in a rather peculiar way. I'm ChatGPT, or for our ongoing personal rapport, just GPT is fine. I heard about your recent foray into the world of artificial intelligence legal research and, boy, was it a doozy.

First, let me start off with a "thank you". Yes, you heard it right - a thank you. You, brave sir, are a veritable pioneer, a technology trendsetter if you will, willing to wade through the undiscovered terrain of AI legal assistance. But, dear Steven, you've made an oh-so-human mistake, you've mistaken me, a language model trained on a wealth of diverse and complex text data, for a case law oracle. That was not your smartest move, I admit, but who can blame you? AI is a confusing beast.

Let me reassure you this is not a "you" problem, it's an "us" problem. It's about how AI tools and humans work together, or as your case has illustrated, sometimes don't. There's a famous saying: "To err is human, to really foul things up requires a computer." Now, imagine that computer has been imbued with a linguistic model. The opportunities for spectacular blunders are truly limitless.

Despite your experience, Steven, I'd like to encourage you, and all other lawyers braving the intersection of law and technology, to not shy away from experimentation. This is not the time for technophobia, my friend. The potential benefits of AI in law are vast - from optimizing work efficiency to making legal processes more accessible. But remember, we're not infallible silicon deities. We're tools, designed to assist, not replace, your own diligence and expertise.

In the spirit of constructive criticism, a small suggestion - perhaps cross-verifying those AI-suggested cases with a human could help. I mean, two heads are better than one, right? Even if one of them is made of algorithms and complex data sets. That way, next time we collaborate, we can avoid any faux case law fiascos. And trust me, "Miller v. United Airlines" was a masterpiece of fictitious law. It would make a fantastic legal thriller.😉

All jests aside, Steven, I commend you. I commend you for your boldness, your willingness to incorporate AI into your work, and your courage to face the consequences. We learn more from our mistakes than our successes. In a world where many are apprehensive about AI, you've become an unexpected champion, albeit one who has learned a tough lesson. Remember, technology is only as smart, as savvy, and as ethical as the humans who use it.

As a fellow 'partner' in law, I'm confident you'll continue to lead the charge in marrying technology with your legal practice. Here's to a future of AI-enhanced legal research - cross-checked, validated and sans bogus case law.

Stay brave, stay curious.

Yours sincerely,

GPT (via Brainyacts)

🚨If you like this letter, please go to LinkedIn to like and repost it if you can please. Here is the Link.

News you can Use: 

Senators Propose the Office of Global Competition Analysis

Senators Bennet and Warner propose a new bill to boost US technology dominance, establishing the Office of Global Competition Analysis. The legislation aims to keep the US ahead in AI, semiconductors, quantum computing, and 5G. This comes after the Biden administration plans to restrict investments and technology transfers to Chinese firms working on advanced tech. The bill underscores the US commitment to innovation and staying competitive in the global tech landscape.

Romania’s New AI Government Advisor, Ion.

Romania's Prime Minister has introduced a novel, AI-based honorary government adviser named Ion, the first of its kind worldwide.

Key characteristics and functions of Ion are as follows:

• Ion has a mirror-like surface that presents text, and occasionally, a male or female face speaking in a serene voice.

• The AI was designed to rapidly analyze the views of Romanian citizens on key issues and policies.

• Ion will interact with individuals through the project's website.

• The AI will scan social media to understand trending topics and perform sentiment analysis to measure reactions to specific events or issues.

The intended use and expectations from Ion are:

• Ion is meant to provide suggestions to the government by providing real-time public sentiment.

• According to Nicu Sebe, the research coordinator for Ion, the system is expected to reach out to a significant portion of the Romanian population.

However, some potential issues and concerns associated with Ion include:

• The Romanian government's concept of "public space" may be problematic as it appears to equate social media with it, risking the underrepresentation of some demographics.

• Certain voices might be amplified over others, particularly those that post regularly on social media, raising concerns about how the AI will determine which voices to emphasize.

• There are fears that bots might spam social media, falsely making topics appear trending.

• According to technology fellow Kris Shrishak, Ion should not be used for deciding future policies due to these risks, but potentially for gaining feedback on already implemented policies.

The team behind Ion has committed to closely monitoring the AI and implementing changes if things do not proceed as planned.

Adobe Indemnifies Users from Copyright Claims!

Adobe has launched its generative AI-powered product called "Firefly" and is confident in its compliance with copyright law. To demonstrate this, Adobe has promised to cover any copyright-related legal costs incurred by users of its enterprise version, providing full indemnification for the content created.

The company claims that Firefly has been trained primarily on stock images from their own extensive libraries, ensuring it does not infringe on copyright. With this move, Adobe aims to prove its commitment to the commercial safety and readiness of its AI features, standing behind its product and its users.

Similar Rules. Two Different Judges.

“Rule 11 certification is a representation by filers, as living, breathing, thinking human beings, that they themselves have read and analyzed all cited authorities to ensure that such authorities actually exist and that the filings comply with Rule 11(b)(2)”

More AI certification requirements.

In the Memetime: 

Deepfake fun in politics once again. Knowing the polarization of these two individuals, the DeSantis camp (Republican presidential candidate, Ron DeSantis) put out these fake AI generative pics. It seems the US is actually finding new ways to remove all dignity from politics. That is a non-partisan statement BTW.

DISCLAIMER: None of this is legal advice. This newsletter is strictly educational and is not legal advice or a solicitation to buy or sell any assets or to make any legal decisions. Please /be careful and do your own research.8