- The Copilot
- Posts
- Deep Dive Analysis of AI Senate Judiciary Committee Meeting
Deep Dive Analysis of AI Senate Judiciary Committee Meeting
Hello 👋
This is The AI Copilot - We’re the substance-over-hype AI newsletter.
You get thoughtful analysis, not links to other products and news.
It’s Wednesday, May 24, and we have six fascinating topics to cover.
😅 AI Fear & Greed Index
📰 Deep Dive - Senate Judiciary Committee Meeting
📚 AI Time Saver - Create and Enrich Tables
👏 AI Research & Products
🎨 Can’t believe human + AI co-created this :)
🌶️ Spicy AI tweets
😅 AI Fear & Greed Index around May 24, 2023
Your weekly barometer1 of global sentiment towards AI, capturing our hopes and fears on a scale from 0 (all dread) to 100 (pure gold rush).
👍 The Good - AI model spots those at the highest risk for up to three years before diagnosis (Harvard Medical, May 8, 2023)
👎 The Bad - Market dips briefly after AI image of fake explosion near Pentagon goes viral (The Hill, May 22, 2023)
🤨 The Interesting - The First Dark Web-Trained AI Is Here (Spiceworks, May 22, 2023)
1 This index is created by analyzing hundreds of content pieces using natural language processing.
📰 Deep Dive - Senate Judiciary Committee Meeting
Last week discussed the Senate Judiciary Committee meeting and highlighted The Hits, The Misses, and Reading Between the Lines.
This week we go a level deeper.
We use a custom-trained AI model to analyze the full transcript using the famous Plutchiks Wheel of Emotions.
This vibrant model dissects emotion into eight fundamental types and their varying intensities.
The eight emotion types are: Joy, Trust, Fear, Surprise, Sadness, Disgust, Anger, and Anticipation.
When applied to transcript analysis, Plutchik's Wheel becomes a sharp decoder, slicing through words and phrases to reveal underlying feelings.
Plutchik’s Emotion Wheel is like an emotion thermometer, measuring the heat and cold of sentiments. This helps grasp the true emotional tone of dialogue, making it an unparalleled tool for conversation analysis.
Here is the analysis:
Total words: 28,300+ words
Democrats: 10,300+ words
Republicans: 4,300+ words
Industry (Sam Altman, Gary Marcus, Christina Montgomery): 13,600+ words
Below each person’s name, there are key emotions that were reflected in their conversation.
⏱️ AI Time Saver - Table Cleanup and Enrichment 🧹
This week, our magical carpet ride in the realm of AI takes us to a place where you'd never think to venture on your own:
The fascinating world of tables.
Yes, you heard right, tables!
Those mundane grids of rows and columns are about to get an AI makeover, turning into treasures of data at your fingertips.
Here's how.
Picture this: you're scouring the web for the world's happiest countries.
You land on a CNN article that lists 20 countries brimming with joy.
You could painstakingly copy each country, paste it into a table, and then spend hours searching for more information.
Or, you could harness the power of ChatGPT.
Copy the list of countries, and with a well-structured prompt (below), ask ChatGPT to not only neatly arrange the information into a table but also to enrich it with additional information for each country.
It's like having a personal research assistant who doesn't need a coffee break.
And voila! Your enriched table is ready, giving you more time to pack your bags for a trip to one of those happiest countries. ✈️
So, this week's AI Time Saver challenge is to test this out.
It's time to give your Ctrl+C and Ctrl+V a rest and let ChatGPT do the heavy lifting.
Let the joy of hassle-free data management brighten your day, just like those happy countries on your list.
After all, who knew data could bring such happiness?
Below is a list of countries.
Take the list and create a table with 7 columns.
Column 1: Serial Number
Column 2: Country Name
Column 3: Capital
Column 4: Population of the Country
Column 5: Per Capita Income
Column 6: Life Expectancy
Column 7: 1 Unique Fun Fact about the Country
Here is the list of countries:
"""
1. Finland
2. Denmark
3. Iceland
4. Israel
5. Netherlands
6. Sweden
7. Norway
8. Switzerland
9. Luxembourg
10. New Zealand
11. Austria
12. Australia
13. Canada
14. Ireland
15. United States
16. Germany
17. Belgium
18. Czech Republic
19. United Kingdom
20. Lithuania
"""
A few things to note:
OpenAI says it is a best practice to provide input in 3 double-quotes - “““<data>“““</data>
Provide clear instructions. We list column-by-column what information we want ChatGPT to return
Keep in mind that the ChatGPT model training data is until September 2021. So this information will be slightly dated.
Here is the ChatGPT response I received based on the prompt.
🙌 Quick request
This newsletter took me 15+ hours to create. If you found this helpful, I’d love for you to share it with a friend.
👏 AI research and products that caught my eye!
Nvidia is researching methods that can simulate tens of thousands of hairs in high resolution and in real time using neural physics. Neural physics is an AI technique that teaches a neural network to predict how objects move in the real world.
Source: Nvidia
Adobe Firefly professional quality content and is designed to be embedded directly into creators' workflows within the Adobe suite.
Firefly is the new family of creative, generative AI models that currently focus on image and text effect generation
Source: Adobe Firefly
🎨 Human + AI co-created this :)
A human created the prompt. AI did the painting. 🤝
🌶️ Spicy AI Tweets
We're falling for this (shit), again:
--Altman, CEO of OpenAI calls for US to regulate artificial intelligence (BBC) May '23
--Zuckerberg, We need a more active role for governments and regulators Oct. '20
--Facebook COO Sandberg calls for government regulation (CNN) June '19— Scott Galloway (@profgalloway)
3:14 AM • May 18, 2023
Ever imagined what famous figures from history would say about AI?
What if they were podcasters?
Here is a fun thread featuring 10 legendary figures and their hypothetical thoughts on AI.
Albert Einstein to kick off this thread 🧵
👇Next up: Mahatma Gandhi as Podcaster
— Salil Sethi (@salilsethi)
5:12 AM • May 24, 2023
“I’m skeptical of the hype around Artificial Intelligence. Conventional intelligence is quite good”
~ Charlie Munger
— Sieva Kozinsky (@SievaKozinsky)
3:02 PM • May 6, 2023
Few things are harder to predict than the ways in which someone much smarter than you might outsmart you.
That's the reason I worry about AI. Not just the problem itself, but the meta-problem that the ways the problem might play out are inherently hard to predict.
— Paul Graham (@paulg)
9:56 PM • May 21, 2023
That’s it for this week. See you next week with another issue of The AI Copilot.
Cheers,
Salil