AI Weekly Malaysia

Simon Willison comments on EmbeddingGemma 2 being under Apache 2.0, arguing that embedding models should not be closed, hosted-only services because apps store thousands to millions of vectors and a vendor deprecation can force costly re-embedding. He notes OpenAI once offered to cover re-embedding costs in April 2024 but says that cannot be relied on, and says he prefers paying a hosted provider while knowing he can fall back to open weights or another vendor.

Why: If you build RAG or semantic search, this is a warning to pick embedding models with an open-weights fallback or multiple hosts, because a model retirement can turn into a full re-embedding bill across your stored vector corpus. EmbeddingGemma 2's Apache 2.0 license gives one such fallback path, but the text gives no benchmarks, pricing, or migration tooling, so it is not a performance or cost recommendation.

Ars Technica published a piece headlined around a survey finding that licensing costs are pushing about 90 percent of VMware users to explore alternatives, with operational complexity cited as a top barrier to migrating. However, the extracted article text contains only cookie-consent and privacy-policy boilerplate — no survey methodology, sample size, vendor names, pricing figures, dates, or quotes are present in the source text.

Why: There is not enough in this text to act on: the only concrete claim is the headline's 90 percent figure and the framing that operational complexity is the top migration barrier, with no respondent count, survey author, or cost numbers to check. If your team is weighing a VMware renewal, treat this as a pointer to go find the actual survey rather than as evidence for a migration decision — you cannot tell from this excerpt whether 90 percent means 'evaluating' or 'already leaving'.