Why can't computers properly add decimal numbers?

Computers use binary floating-point representation (usually IEEE 754 standard), which cannot precisely represent many decimal fractions. For example, the number 0.1 in binary is an infinite repeating fraction, so it is stored with some approximation.

Because of this, small rounding errors can occur when adding decimal numbers. This is not a bug, but a feature of how numbers are represented in computer memory.

To avoid such issues, in tasks where decimal precision is important (e.g., finance), special data types are used — decimal arithmetic or libraries for fixed-point calculations.