What Drives AI Policy Impact? A Data-Driven Look at 30 Key Policies

By shrijeetverma13 · July 20, 2026

This analysis examines 30 major AI policies across four categories—Regulation, Corporate, Diplomacy, and Trade—to identify what makes some policies more…

A bar chart and data table were generated showing the average policy impact score for each policy category, ranked from highest to lowest. This lets you quickly see which categories have the strongest average impact.

The analysis grouped policy impact scores by the month articles were published and calculated the average score for each month. A bar chart was generated showing how average policy impact scores rise and fall across different published months, along with the number of records in each month.

The analysis generated a line chart showing how the volume of published policies has changed month by month, along with supporting data tables. The visualization plots the number of policies published in each month, making it easy to spot upward or downward trends, peaks, and slow periods over time.

Out of 30 total policies, they are spread across 4 categories: Regulation leads with 20 policies (66.7%), followed by Corporate with 8 policies (26.7%), while Diplomacy and Trade each have just 1 policy (3.3%). A bar chart and pie chart were created to visualize this distribution, along with a data table.

The analysis identified the top 10 policy titles with the highest policy impact score, displayed in a horizontal bar chart ranked from lowest to highest. The chart shows each policy's title alongside its impact score, colored by score intensity, making it easy to spot which policies rank at the very top.

Yes, there is a big difference. Policies with missing summaries have a much lower average policy impact score (0.0) compared to policies that have summaries (4.88), a difference of 4.88 points.

A summary table and grouped bar chart were generated showing the average policy impact score for each category, along with the minimum, maximum, and count of entries per category. The bar chart visually compares the average impact scores across all categories, making it easy to see which categories tend to have higher or lower policy impact.

Your articles were grouped by month and policy category, and a stacked bar chart was created showing how the mix of categories like Regulation, Diplomacy, and others has shifted over time. The chart and accompanying data tables give you a clear month-by-month view of category distribution.

The analysis extracted the top-level domain from each article's source URL and ranked news outlets by both article count and average policy impact score. Two charts were generated: one showing the top 15 sources by number of articles published (colored by their average impact score), and another showing the top 15 sources ranked by average policy impact score (colored by article volume). Together, these visualizations highlight which domains are the most prolific publishers of AI policy news and which ones tend to cover the highest-impact stories, even if they publish less frequently.

Using the IQR method on the policy impact score column, no outliers were detected. The first quartile (Q1) was 2.25 and the third quartile (Q3) was 6.00, giving an IQR of 3.75. This sets the acceptable range for typical values between -3.38 and 11.62. Since every entry's policy impact score fell within this range, there are no unusually high or low scores to flag — the dataset shows a fairly consistent and well-behaved distribution of impact scores across all titles and categories.

The sentiment analysis of the 'summary' text column shows that most entries (66.7%) are neutral in tone, while 23.3% carry negative sentiment and only 10% are positive. The average polarity score of -0.014 suggests the overall tone is very slightly negative but close to neutral. A bar chart visualizing this distribution has been generated.

A word cloud was generated from the 'summary' column, visually representing the most frequently occurring words in the text data. Larger words indicate higher frequency, making it easy to spot dominant themes and topics at a glance.