Implementing guardrails against malicious use by those in power is sufficient to safely continue artificial intelligence development