In fact, AAE uses be to mark a habitual grammatical aspect, which is not explicitly distinguished in Standard English.
Some linguists suggest it came from the finite be in the 17th-to-19th century English of British settlers (perhaps especially those from South West England, but the usage may be the recent "Mummerset" in this context).
Evidence includes the fact that both dialects structure sentences with the habitual be almost identically: (1) Even when I do be round there with friends, I do be scared.
(AAE)[4] Criticism of that hypothesis stems from the fact that there is no evidence that be has been used as a habitual marker either in the past or today in Caribbean creoles of English.
Although the expansion to include the differences in the dialects of HE accounts for the absence of habitual be in Caribbean English creoles, the hypothesis has its disadvantages as well.
In America, the Irish feature habitual be may have diffused into AAE and the two assemblages of people were in close contact and communicating with a new tongue.
It is possible that British dialects could have had features that served as models for habitual do (be) in the Caribbean creoles, which, in turn, expanded to AAE in the Americas.
It is also worth noting that Southern and Southwestern British immigrants traveled to the American colonies as well and their dialects would have been used as a model to Black people, leading to a drawback, the introduction, and subsequent loss of habitual do (be) in America, which was also the problem with the first expansion of the diffusion hypothesis.
Rickford gives evidence: (18) Habitual aspect with a prepositional phrase or locative: Stage 1: He (d)a de [dc] in the bed.
[4] Another merit is that this same pattern of decreolizing of be is found in other creoles that are relatively close to AAE and affirm the plausibility of this origin for habitual be.