AI at Work and in Public Life: Scribes, Code, and Deepfakes Redefine Trust

Artificial intelligence has slipped from the lab into the fabric of everyday work and public life. In clinics across Australia, a rising share of GPs now use AI scribes to record conversations with patients, a shift that prompts a key question: are patients and doctors trading care for convenience? Reports show that about two in five Australian GPs use AI scribes, which some clinicians say frees them to connect more fully with patients, while others warn it could erode the human touch if consent isn’t clear and if the technology misreads nuance. This tension mirrors a broader pattern seen across sectors as AI becomes a daily partner rather than a distant tool.

Momentum is equally evident in software development. A leader who moved an entire team to an AI-first workflow describes a dramatic acceleration: throughput leaps to roughly 170% while headcount falls to about 80%, and quality assurance evolves into a system of automated validation. Ideas no longer languish on whiteboards; they can morph into a working prototype, a PRD, a tech spec, and a production-ready feature within days. The result is a shift from a rigid diamond-shaped process to something closer to a control tower, where humans set intent and constraints, AI executes with speed, and verification closes the loop before production.

In practice, that means validation becomes inseparable from creation. The organization now uses AI to generate tests, enforce coverage, and maintain guardrails. QA specialists have become architects of the system, building AI agents that convert requirements into automated acceptance tests, and the organization measures success by how reliably AI produces correct outcomes at scale. It’s a reframing of what “shift left” means: correctness is baked into the production process rather than tacked on after coding.

But the rise of AI also amplifies risks in data and information ecosystems. In one widely discussed case, fraudulent church data exposed vulnerabilities in polling, where paid participants deploy automated tools to generate unreliable responses. The result is a critical reminder that the integrity of data informs policy, democracy, and public trust—areas where AI acceleration can backfire if the inputs are themselves compromised or manipulated.

Public messaging and policy communication are not immune. A recent AI-generated video from a US embassy in Mexico urging migrants toward “self-deportation” sparked outrage online, underscoring how quickly synthetic media can shape perceptions—whether intentional or not—and how difficult it is to police misused AI in official communications.

And beyond the policy arena, political deepfakes and AI-driven branding prompt new questions about influence and authenticity. Research points to the spread of convincing AI images and avatars used for propaganda or monetization, while opinion pieces question the rising prominence of CEOs in the narrative—branding exercises that can feel transactional in a world where AI and media blur the lines between message and messenger.

Taken together, these stories sketch a future where AI shifts not only what we can do, but how we think about work, trust, and governance. The challenge is to preserve human oversight and ethics while unlocking AI’s promise: to augment judgment, speed experimentation, and improve outcomes—without eroding consent, accountability, or truth. As readers, organizations, and policymakers, the path forward is to embed responsible AI practices, invest in cross-functional validation, and anchor innovation in empathy for the people who will be affected.

Sources

1. The Guardian: Two in five Australian GPs use AI scribes to record patient notes
2. When AI turns software development inside-out: 170% throughput at 80% headcount
3. ‘Our assumptions are broken’: how fraudulent church data revealed AI’s threat to polling
4. US embassy in Mexico prompts outrage with AI video promoting ‘self-deportation’
5. ‘They feel true’: political deepfakes are growing in influence – even if people know they aren’t real
6. These CEOs want a starring role in our lives – and there’s not much we can do about it | Larry Ryan