Evidence, with its limits intact
These examples show what deliberate AI adoption can enable. Each claim links to an official source and is clearly separated from what another organization should expect.
Wix acquired Base44, an AI-powered platform for creating software applications through natural-language instructions, for initial consideration of approximately US$80 million plus potential earn-out payments.
Strong results come from a defined workflow, suitable data, human review and measurement rather than AI access alone.
Reported by Wix. Results are organization-specific and are not a promise of equivalent outcomes.
Read the official source ↗OpenAI documented a five-month engineering experiment in which Codex generated every line of application logic, tests, CI configuration, documentation, observability and internal tooling.
Strong results come from a defined workflow, suitable data, human review and measurement rather than AI access alone.
Reported by OpenAI. Results are organization-specific and are not a promise of equivalent outcomes.
Read the official source ↗During Alphabet’s February 2026 earnings call, its CFO stated that about 50% of Google’s code was being written by coding agents and then reviewed by engineers.
Strong results come from a defined workflow, suitable data, human review and measurement rather than AI access alone.
Reported by Alphabet. Results are organization-specific and are not a promise of equivalent outcomes.
Read the official source ↗AWS reported that Amazon Q Developer assisted with migrating tens of thousands of production applications, saving more than 4,500 developer-years of work and producing performance improvements worth approximately US$260 million in annual savings.
Strong results come from a defined workflow, suitable data, human review and measurement rather than AI access alone.
Reported by Amazon Web Services. Results are organization-specific and are not a promise of equivalent outcomes.
Read the official source ↗Morgan Stanley reported adoption by more than 98% of advisor teams and substantially broader access to internal knowledge through its AI assistant.
Strong results come from a defined workflow, suitable data, human review and measurement rather than AI access alone.
Reported by Morgan Stanley. Results are organization-specific and are not a promise of equivalent outcomes.
Read the official source ↗Moderna reported more than 750 custom GPTs, with 40% of active users creating GPTs and an average of 120 conversations per week.
Strong results come from a defined workflow, suitable data, human review and measurement rather than AI access alone.
Reported by Moderna. Results are organization-specific and are not a promise of equivalent outcomes.
Read the official source ↗