Wolfram Alpha's math input is broken

There is a serious bug in Wolfram Alpha's "math input" mode. When you enter e²ⁿ, it is interpreted as e²n (full details at the end). This was reported to them a month ago and still hasn't been fixed, so I figured it was time for some public shaming ;)

I've been really impressed with Wolfram Alpha over the years (both the natural language parsing and the power of Mathematica); my main issue until now has been that the natural language parser tends fail on inputs beyond some length (fortunately Mathematica syntax is also supported and works well). So I was very surprised when this glaring bug in math input mode was shared with me.

Full steps to reproduce:

1. Go to https://www.wolframalpha.com/

2. Click "MATH INPUT"

3. Click the "power" button (second from the right, icon is two boxes with one in superscript)

4. Type "e" (it should go in the first box)

5. Click the superscript box

6. Type "2n"

7. Click the "=" button

Result: The input field correctly shows e²ⁿ (with the "n" in superscript), but the formula shown in the "Input" section is e²n (the "n" is outside the exponent) and the "Plot" section shows a straight line which confirms that the input was misinterpreted as e² * n.

Explicitly adding parentheses around the "2n" fixes this. Ironically, when you do that the "Input" section shows the formula as e²ⁿ (without the parentheses; the same version that fails when entered in the input field).

Wolfram Alpha's math input is broken in other ways as well: https://bostick.github.io/blog/2024/05/precision-bombing/

Entering input such as: 1*^-1355718576299609

or:

1.0`90071992

will trigger bad behavior on the site.