As a general effect evident for the population overall, it is not true that IQ's / 'intelligence' (as some sort of definitive factor(s) or capability) have been demonstrated to be universally increasing.
As an effect evident within more specifically defined sub-populations (as delimited geographically; demographically by cohorts; by age; by timeframe; etc.) it is true that IQ
scores (ratings derived from normed IQ test instruments) have been shifting.
The general label for this phenomenon is the Flynn Effect:
https://en.wikipedia.org/wiki/Flynn_effect
This was first noted circa 40 years ago, when it was discovered that later groups taking older IQ tests scored higher (as a group) than groups who'd taken the old tests back when they were the latest / current standard.
It's critical to bear in mind that IQ scores are relative rankings mapped onto a presumed normal (Bell Curve) distribution. It's the collective test performances of the entire group that are used to establish the mean / average, after which individuals' scores are plotted in relation to the mean thus established.
It's generally accepted that there's been a positive shift in the mean for certain populations of children for which there's sufficient data to make comparisons. The situation is much murkier for adults.
It's also generally accepted that this positive shift - if / when evident - is probably influenced by improvements in (e.g.) general health, general nutrition, more cognitive / intellectual stimulation occurring earlier, and improved access to - and quality of - educational opportunities.
Speaking broadly, and solely in the context of those studies based on substantial evidence ... The shift isn't universally apparent, and it's not uniform across the range of IQ scores. There are cases where IQ scores for a given population or sub-population have been stable or even declined. The bulk of the positive shift seems to consistently pertain to the lower reaches of the scores' range. The most apparent upward shift occurs among the lower scores, and the most apparent non-shift or decline occurs among the higher scores. Phrased another way - the most demonstrable aspect of the rising mean / average is that it represents improved performance in the 'relatively low' range rather than improved performance across the entire range. Phrased even more simply - the mean is floating upward because the lower scores are trending higher, not because all scores are trending higher.