26th Anniversary of the biggest shake up in ITV

I'm not confident that this was the reason. I think the basic summary of what happened in the countries that used NTSC was that commercial broadcasters had to work with whatever TVs were used in the country and they didn't have teletext decoders installed. If a situation had emerged where over half of all TVs sold in the late 1980s had teletext decoders then it's possible that some broadcasters would have tried out teletext services in one form or another as the receiving infrastructure is already in place.

Brazil is an oddity when it comes to TV standards and is also a country with a large population. Does anybody know anything about the PAL-M TVs used in Brazil, such as which companies manufactured them or whether they are unique designs or modified NTSC TVs etc?

The cost to a TV company of implementing a (basic) teletext service is quite low. It doesn't necessarily have to be anything as grand as Ceefax or Oracle.

Teletext is quite strongly tied to PAL (apart from PAL-M) as it was developed in Britain. Therefore countries using PAL were in luck when it came to deploying teletext services because PAL TVs with integrated teletext decoders were mass produced commercially available products in the 1980s.

Countries using NTSC and SECAM had to develop / adapt teletext to their systems. They were less successful either at adapting teletext or manufacturing TVs incorporating teletext decoders in large quantities. France and Japan were the most successful countries.

What happened in Russia? I'm led to believe that there were experiments with Antiope and the SECAM used in communist countries during the early 1980s but some time later (when?) Russia and some other CIS nations adopted the WST teletext standard but it cannot display Cyrillic characters.

RUssian and other Cyrillic characters are part of Teletext ..
See table 39 of EN 300 706
https://www.etsi.org/deliver/etsi_i_ets/300700_300799/300706/01_60/ets_300706e01p.pdf

Teletext supports Latin characters with 13 different language sets plus X/26 diacritical marks
But only English has @
And then Cyrillic Greek Arabic and Hebrew .

WST I.e the UK DTI specification had ideographic including Manderin and Kanji
But the EBU and then ETSI did not perpetuate them !!!!

SECAM As has been noted had the problem of the bottles in the VBI ...

Do the teletext chipsets in Russian TVs support this standard because the Russian teletext I have seen is in the Latin alphabet?

These days there aren't likely to be separate chipsets for different character sets - and multi-character set chipsets will be the norm. I think this has been the case since the 90s. (Cheaper to have one multi-standard design than different designs for each country - both in TV and chipset terms)

It may have been that early WST transmissions didn't transmit a character set flag and/or early receivers only had one character set in ROM (ISTR that German teletext on Astra 1 broadcasts displayed accented characters incorrectly on the late 80s TVs I was watching on) - so you had to have the correct chipset for display, but with the move to satellite TV and pan-European model designs, I'm pretty certain that the current spec includes character set flags (and has done for a while), and most receivers will cope with multiple character sets.

I know my Sony Level 2.5 (at least) decoder copes fine with multiple character sets (I've fed it all sorts of different nationality teletext content via both DVB-T and Analogue VBI and it's correctly handled it. It's also nice having more than 7 colours and less-blocky graphics when you feed it Level 2.5 stuff )