The Wonder and Victory-Point VCs are
always turned off by default in the epic game — both these VCs are more intended to be used for (time-limited) Multiplayer games, or for specific Scenarios (e.g. the 'Mesopotamia' Conquest-scenario uses both the Wonder-VC, and VP-scoring). So no, turning the Wonder-VC off (or rather,
leaving it off) will
not discourage the AI-Civs from building Wonders!
A VP-win is achieved by accumulating a preset number of VPs, and AFAIK, even if VP-scoring
is turned on for the epic-game, VPs do
not directly contribute to the Histo VC, per se. The obelisks you see when you turn on VP-scoring are the 'Victory-Point locations' (on a randomly generated map, like the Strategic resources, the VPLs will also be randomly scattered): for every turn that a unit occupies a VPL-tile, the unit's parent Civ accumulates XX VPs. The actual VP-value awarded for this and various other conditions (e.g. no. of cities founded, no. of enemy units killed/captured, no. of Wonders built, etc.), plus the 'total-VP' target, are all set in the
[Scenario-name].biq file, but off the top of my head, I don't know if VP-targets and values were
also set for the epic-game
conquests.biq file (if you open it in the Editor, you can check for yourself). If so, then it is
possible that a VP-win might be (accidentally) achieved by you or one of the AI-Civs — likely largely depending on difficulty level — before the timer runs out for a Histo-VC.
I have no idea what would happen if you turned off the Cultural VC(s) — which
are part of the default epic-game — but I suspect it would not make a significant difference to the AI's behaviour, since the Civ3-AI does not 'plan' to achieve
any specific VC to begin with: when it does so, this happens entirely by 'accident'. Turning off the Cultural VC will certainly not discourage the AI from putting up Culture-producing buildings, since Wonders, Temples, Libraries, etc., all provide additional benefits apart from Culture. Anyway, if an AI-Civ is left to its own devices for long enough, it will tend to build everything it can, in every one of its towns (without making any kind of cost-benefit analysis). And besides, the actual conditions required for a Cultural VC are not very conducive to the AI winning that way: at Regent and below, the human is more likely to do that (by snaffling all the Wonders), but at Monarch and above, the AI's tendency towards militarism generally means that Civs with good lands/ strong traits will simply roll over the weaker Civs, rather than beating them Culturally.
The 20K-City VC requires an extremely Wonder/Culture-focussed approach for just one city, which the AI is largely incapable of following: the AI-Civs instead tend to spread their Wonders around multiple cities. To win a 100K-Civ VC requires that the
next-best Civ still has <50K Culture, but again, due to the AI's hard-coded tendency to found towns 'looser' than the human (Cxx[x]xC rather than Cx[x]C), AI-Civ-wide Cultural production rates tend to be quite slow, and 100K Culture usually isn't achieved by any AI-Civ until the late Industrial. But that far through the game, even if the human player is significantly behind on Cultural production (which will usually be the case, for a long game where the human has rather focussed on Science and/or Military), the 2 largest AI-Civs still standing will likely be
relatively equally balanced (due to their tendency to build everything everywhere), so neither of them will pose much if any 'danger' of winning that way, either.
FWIW, I have
never seen an AI win by Culture in any of my epic-games (usually played on Continental/Pangaea maps, usually at Emperor/DG these days), but
it apparently does happen occasionally!