At least three inventors have a real claim, by different tests. Thomas Saint patented a design in England in 1790, but it was never built and stayed unknown until 1874. Barthelemy Thimonnier patented a working machine in France in 1830 and ran a small factory of them until rioting tailors, fearing unemployment, destroyed it in 1831. Elias Howe patented a lockstitch machine in the United States in 1846, sixteen years after Thimonnier, but his patent was the one that survived litigation and anchored the industry America and much of the world adopted. Whether invention means the first patent, the first working machine, or the first patent that mattered commercially is a genuine, unsettled question, not a gap in the record.
What would resolve this A single, agreed test for what invention means here: patent priority, working-machine priority, or commercial consequence. Historians of technology have not converged on one, and probably will not, since the three tests answer different questions.
OpenHistory of technologyElias Howe (Wikipedia)