You can't compare a modern word with a modern word, which is how a lot of the Western linguists, often not even understanding the languages, analyze.. by going to a dictionary.
Yeah, but actually like in Tibeto-Burman you can often find clear cognates. Words like "I" are clearly related even in modern forms. In fact for Thai it's closer to Miao-Yao, and Austronesian.
And what I use are so-called Yakhontov's most-stable meanings, though actually some of them are related to other meaning and some of them can be considered "not very basic".
To show that Tai-Kadai is far away from Sino-Tibetan, I think I can compare with Austroasiatic....
[I also see Viet-Muong seemingly have more blanks in "Austroasiatic etymology" compared to some other, but that doesn't disprove, at least the link is much better than Thai to Old Chinese.]
I'll compare Monic, reconstructed Old Chinese, Viet-Muong, while I know the wordlist may not be complete.
I tried to link Tai-Kadai and Sino-Tibetan but it's rather frustating. There's no "Miao-Yao" etymology in Starostin's site but my guess is that it "scores better" (closer to Sino-Tibetan) than Tai-Kadai or Austroasiatic.
Anyway looking through the database I don't see the reason for "Austric" to exist well as so few words are identifiable. Some linguists are too clever

Example, "egg" in Tai is "kai". That's different from "egg"/"dan" in Mandarin. But, maybe the Tai "kai" is cognate with the Mandarin "ke", which means "shell" or something with a hard outter finish.
Another example, "sawn" is "two" in Tai but "two" in Sinitic is "er/yi/ni", though Shanghainese/Wu uses "liang". But "shuang" in Chinese means a pair.
Yeah, Starostin was clever enough to make those cross-meaning relationships, then the product was Sino-Caucasian

Anyway, according to him, Chinese is closer to Caucasian than Tai-Kadai

He was a better linguist than you and me
Oh, I do believe Tai-Kadai and Sino-Tibetan are related. One day I hope to see it proven.
In the big picture I posted (post#26), every non-African non-Melanesian meets at single point

It's actually more like "song". There are also cases where "yi" is used. Like sometimes the second month is called "duan yi", where duan just means moon or month. When counting in Tai and you reach number 20 it is called "yi sip", where yi is 2 and sip is 10. The number 21 is "yi sip et". In this case, the et just means first, rather than 1. The number 22, is "yi sip song" where obviously song is used as the second 2.
Using of foreign numerals are proven in Japanese and Korean....
I've also had my doubts on linguististic classification. It seems so arbitrary. It seems to be based on the preconcieved notion that languages sprang up cleanly and in an orderly fashion. Maybe a language came to be because it was a fusion of 3 or 4 languages? Who knows.
No, the base is that languages never loan too many basic words. It could be replaced with another language with some "original language" loans. In English, Latin have also mixed in but the core words are still recognizable.
Some are called "mixed Language": Wutunhua is Chinese-Tibetan-Mongolian. Looks like rather modern mix