# Subtle Logic Bugs — AI Coding Assistants: Use Them Well and Review Them Hard

Source: https://www.geekswithgeeks.com/en/ai-coding-assistants/rev-logic-bugs

> Catch off-by-one errors, missing edge cases and shared state mistakes in generated code.

## Right shape, wrong detail

Generated code often has the right structure but a wrong detail: an **off-by-one** error, a missing check for an **empty list**, a **boundary** that should be inclusive, a **shared mutable default**. These pass a casual glance. Run the code on boundary inputs (0, 1, empty, the maximum) and read loops and indexes slowly.

## Two classic slips, run

I ran this. `average([])` raises ZeroDivisionError. `page_bad` treats page 1 as the second page, so it skips the first three items; `page_ok` uses 1-based pages.

```python
def average(nums):
    return sum(nums) / len(nums)        # crashes on []
def average_safe(nums):
    return sum(nums) / len(nums) if nums else 0.0

items = list(range(1, 11))
def page_bad(items, page, size):
    return items[page * size : page * size + size]     # 0-based by accident
def page_ok(items, page, size):
    start = (page - 1) * size
    return items[start : start + size]

print(average_safe([]), average_safe([2, 4]))
print(page_bad(items, 1, 3), page_ok(items, 1, 3))
```

Output:

```
0.0 3.0
[4, 5, 6] [1, 2, 3]
```

## Try the zero, one and many cases

Most bugs hide at the boundaries. For any function, test with nothing, exactly one item, and many items before you trust it.

**Quiz:** Which input is most likely to expose a missing-check bug in `average(nums)`?

- [ ] A list of 1000 numbers
- [ ] A list of strings only
- [x] An empty list
- [ ] A very long variable name

*Answer:* An empty list. Empty input is a classic boundary case that triggers division by zero.