[guided]The only point that needs proof is that the new variable $x_{n+1}$ is not already generated by the earlier variables $x_1, \ldots, x_n$. To test this, we use an evaluation map that kills precisely the first $n$ variables.
Fix a positive integer $n$. Define a $k$-algebra homomorphism
\begin{align*}
\pi_n: R \to k[x_{n+1}, x_{n+2}, x_{n+3}, \ldots], \qquad x_i \mapsto 0 \text{ for } 1 \le i \le n, \qquad x_i \mapsto x_i \text{ for } i \ge n+1.
\end{align*}
This is a legitimate homomorphism because elements of $R = k[x_1, x_2, x_3, \ldots]$ are finite polynomials: each polynomial mentions only finitely many variables, so substituting $0$ for $x_1, \ldots, x_n$ produces another ordinary polynomial.
Now take any $f \in I_n$. By the definition of the generated ideal $I_n = (x_1, \ldots, x_n)$, there exist polynomials $f_1, \ldots, f_n \in R$ such that
\begin{align*}
f = \sum_{i=1}^{n} x_i f_i.
\end{align*}
Applying $\pi_n$ to this expression gives
\begin{align*}
\pi_n(f) = \sum_{i=1}^{n} \pi_n(x_i)\pi_n(f_i).
\end{align*}
For every index $i$ with $1 \le i \le n$, we have $\pi_n(x_i)=0$, so the right-hand side is zero:
\begin{align*}
\pi_n(f) = 0.
\end{align*}
Thus every element of $I_n$ is killed by $\pi_n$.
But the next variable is not killed:
\begin{align*}
\pi_n(x_{n+1}) = x_{n+1}.
\end{align*}
The polynomial $x_{n+1}$ is nonzero in the [polynomial ring](/page/Polynomial%20Ring) $k[x_{n+1}, x_{n+2}, x_{n+3}, \ldots]$. Hence $x_{n+1}$ cannot belong to $I_n$, because every element of $I_n$ maps to $0$ under $\pi_n$. Since $x_{n+1}$ is one of the generators of $I_{n+1}$, we have $x_{n+1} \in I_{n+1} \setminus I_n$, proving that $I_n \subset I_{n+1}$ is proper.[/guided]