123 | 🐂 💩What is bullsh_t work?

In today’s Brainyacts:

1. What is bullsh_t work?

2. A judge trolls Claude

3. Poll time (unless you already have taken it)

4. X (Twitter) will use your data to train AI and other AI-related news

5.  GenAI training as a work perk and other news you can use

👋 to new subscribers!

To read previous editions, click here.

Lead Story

🐂 💩 Is GenAI only good for bullsh_t work?

Over the last few days, a debate on the usefulness of generative AI for legal work was catalyzed by a back-and-forth between Brian Inkster, myself, and many others.

• We touched on what is legal work and what is bullsh_t work.

• Where is GenAI making an impact, if it is at all?

• Does the nature of the work matter if GenAI can make any of it better and easier?

This was refreshing as we disagreed but worked to flesh out our ideas and stances. If you don’t know Brian, he has been a vanguard in the legal industry.

But let’s turn to OpenAI to ask it for a better description of who Brian is. BTW this is exactly the kind of bullsh_t work that Brian says GenAI is only good for. “Brian Inkster is the founder and CEO of Inksters, a UK-based law firm. He is known for his forward-thinking approach to law and has been recognized for his innovative use of technology in the legal sector. Brian is an advocate for modernizing legal practices and has often shared insights on the potential and limitations of technology in law.”

Using OpenAI above saved me likely 5 to 10 minutes. I would have had to verify what I knew of him and then succinctly typed it out. Perhaps I should be able to do that on the fly. But here I sit next to a freshwater lake working on this. 5 to 10 minutes gives me some time back. And that is the point of GenAI - one of the many.

Moving along - here is the post from Brian where he calls me and others out (in a cheeky but respectful way). You definitely should read it but for the sake of this post, here is the gis of his POV: There is a lot of hype. There is a lot of magical thinking and talking. And there is a lack of legal-specific use cases.

He is not necessarily wrong. But it is not that simple. As I replied with the following:

• Definition of "Legal" Work

  • The perception of "legal work" is often self-referential and not necessarily extraordinary.

  • There's a tendency to undervalue work that lawyers don't personally handle.

  • The distinction between "bullsh_t" work and essential tasks in running a legal business needs clarification.

• Practical Use of Generative AI in Legal Field

  • Generative AI has real-world applications in the legal sector, blurring the lines between "bullsh_t" and "legal" work.

  Examples include: 

  a. Template Banks: Small firms and solo practitioners use GenAI to adapt existing templates, saving time and ensuring accuracy.

  b. Improving Communication: In-house counsel used GenAI to bridge the communication gap between legal teams and business units, creating content that's both legal and business-friendly.

  c. Specific Legal Tasks: GenAI can assist in drafting consent notices or preliminary responses to proposed MSAs, aiding even experienced lawyers.

• Value of "Bullsh_t" Work

  • Even if some tasks are perceived as less valuable, they are essential for the smooth operation of a legal business.

  • Addressing these tasks efficiently can prevent issues like overwork, stunted growth, and reduced profits.

  • Tools that excel in handling these tasks should be appreciated for their contribution to better business operations.

What say you? Comment on my LinkedIn post here.

Spotlight Story

While I admire this judge’s public experimentation, it demonstrates the fundamental misconception that continues to plague these tools.

1. These are not fact machines.

2. These are thinking tools.

3. They make no promises to be accurate. In fact, they disclaim otherwise (even though these disclaimers are tough to find at times).

4. We do not have a familiar category to relate to these tools - so we are defaulting to - they must be bad because they don’t tell the truth.

5. Don’t use them to find the truth. Use them to think.

If you haven’t taken this poll yet, please take 5 seconds and answer this poll. I will share the results.

Thank you!

AI Model Notables

• X’s (Twitter) privacy policy confirms it will use public data to train AI models

• Microsoft files new AI-powered smart backpack patent. It can hear you and see what you see

• Meet Ernie, China’s answer to ChatGPT

• Google is embedding Google Cloud and AI in more automotives

• Zoom deepens its embedding AI into products and services

🚨 BTW I have heard from a couple of law firms that they now have banned the use of Zoom due to its recent snafu on articulating how it is using data from Zoom calls. They did clarify they will not, but it seems trust has been eroded.

☄️ Shoot me a note if your organization has had a similar reaction.

News You Can Use:

➭ Washington Post Roundup of Global AI Policies

➭ 5 Things CFOs Should Know About Generative AI

➭ ChatGPT might change the college application essay forever

➭ More companies see ChatGPT training as a hot job perk for office workers

➭ Criminals use artificial intelligence in 'fraud-for-hire' commercials

➭ Managing the Carbon Emissions Associated with Generative AI

➭ The UK releases key ambitions for global AI summit

Who is the author, Josh Kubicki?

Some of you know me. Others do not. Here is a short intro. I am a lawyer, entrepreneur, and teacher. I have transformed legal practices and built multi-million dollar businesses. Not a theorist, I am an applied researcher and former Chief Strategy Officer, recognized by Fast Company and Bloomberg Law for my unique work. Through this newsletter, I offer you pragmatic insights into leveraging AI to inform and improve your daily life in legal services.