BOB STEWART · KENNEBUNK, MAINE
For more than three decades, I have worked inside systems where identity, safety, medicine, technology, capital, and human judgment carry real consequence. I now build the leadership and trust architecture that allows intelligence to serve without allowing power to escape accountability.
Technology executive · Systems architect · Founder · Steward
I am interested in what happens when a system meets the world—and someone has to answer for the result.
Across early internet platforms, clinical research, identity, security, biometrics, regulated manufacturing, and autonomous intelligence, the pattern has remained remarkably consistent: capability scales faster than accountability unless both are designed together.
The titles changed. The environments changed. The underlying questions did not.
Who is allowed to act? What are they allowed to do? Who carries the consequence? What evidence remains? Who can stop the system when reality departs from intention?
This is the work beneath all of my work.
Power is not made trustworthy by good intention. It becomes trustworthy through bounded authority, visible consequence, and durable evidence.
Organizations and titles matter here only as evidence of proximity. The more important record is what each environment demanded—and what it made impossible to unsee.
The early commercial internet rewarded speed, but it also exposed how quickly infrastructure becomes society. Building platforms before the rules had settled taught me to look beyond whether a system could scale to what its scale would make possible—and what it would make difficult to reverse.
At CyberMedica, regulated clinical technology made the stakes immediate. A workflow was never merely convenient or inconvenient. Identity, consent, protocol, data integrity, time, and accountability had to remain connected because patients, investigators, institutions, and regulators could all inherit the consequence.
Identity systems taught me that knowing a name is not the same as knowing whether a person is authorized, qualified, current, consenting, or accountable in context. Trust fails when institutions reduce identity to a record and ignore the authority surrounding an action.
VitalLock existed to guard what people entrust to systems that outlive them: identity, records, intent, and the keys protecting all three. When a government demand arrived — escrow the keys, or face takedown — the architecture and the promise became the same question. A protection that can be silently overridden is not protection. I refused the escrow and accepted the consequence.
As CTO of EMC Global Security Solutions, I worked in a breach-era environment where confidence, evidence, institutional memory, and executive judgment mattered as much as technical control. Security was not a product feature. It was the ability to remain truthful and governable when assumptions failed.
Serving as COO and a board director at Sonavation widened the frame from architecture to the whole operating system: product, people, quality, capital, timing, governance, and company-level consequence. A technically correct answer is not enough when the institution cannot execute it.
After decades in technology leadership, I returned to the manufacturing floor and shift leadership inside regulated life-sciences production. Heat, tooling, robotics, materials, maintenance, handoffs, and human fatigue do not negotiate with abstraction. The experience renewed my respect for the people who make systems real and for the disciplines that keep those systems safe under pressure.
Autonomous systems return every earlier lesson at once. Identity is insufficient without authority. Policy is insufficient without enforcement. Action is insufficient without evidence. Intelligence is insufficient without review, revocation, and a humanly accountable chain of consequence.
My present work connects executive leadership, lifecycle governance, institutional review, verifiable authority, and durable evidence so that increasingly capable systems can serve without becoming unanswerable.
Organizations are shown as career environments and do not imply current affiliation with, endorsement by, or sponsorship of Bob Stewart or his present work.
Capability does not grant permission. A consequential system should know who delegated authority, what scope was granted, which constraints apply, and how that authority can be withdrawn before it acts.
A decision that cannot be reconstructed becomes institutional folklore. Identity, policy, evidence, judgment, execution, time, and outcome should remain connected in a durable record.
Good governance is not the brake on responsible innovation. It is the steering, instrumentation, memory, and shared confidence that allow an institution to accelerate without becoming reckless.
Boardrooms, codebases, clinics, incident rooms, rescue scenes, and factory floors each reveal truths the others can miss. Judgment improves when leadership remains close enough to reality to be corrected by it.
Institutions often speak of human resources while designing environments that consume human capacity. Technology should expand agency, dignity, competence, and contribution—not merely make extraction more efficient.
What a system protects, rewards, remembers, permits, and refuses is not incidental. It is the moral structure made operational.
The work is not to slow the future down. The work is to make the future worthy of acceleration.
These efforts are not one product and should not be mistaken for one legal entity. They are connected by a common commitment: intelligence and institutions should be able to prove who acted, under what authority, against which evidence, with what review, and to what consequence.
Executive Technology Leadership
My deliberately small fractional and interim CTO practice for organizations facing consequential technology decisions. MaineSoftware is where executive mandate, authority, staffing, budget, architecture, and outcomes become one accountable system.
Institutional Review and Oversight
The review chamber for consequential systems: who is competent and authorized to examine consequence before deployment, what evidence must be present, and how approval, dissent, holds, incidents, and appeals remain on the record. This includes the work of AI-IRB and OpenIRB.
Lifecycle and Review Governance
Methods, gates, review structures, and education for making AI systems governable from conception through deployment, change, incident, and retirement. This includes the work of AI-SDLC Institute, AI-IRB, IRB.Institute, and OpenIRB.
Clinical and Medical Consequence
The clinical expression of the trust architecture: governance, oversight, evidence, and operational accountability for medical and research systems where failures can reach patients directly.
Personal Safety and Continuity
The most personal expression of the work, rooted in two decades of rescue service and in VitalLock, the digital-legacy platform where I learned what protection must be willing to refuse. LiveSafe carries that lineage forward: safety, identity, and continuity for the people a system exists to serve.
Leaders, Teams, and Contribution
A practical ecosystem for developing technology leaders, shaping opportunities, building governed systems, and recognizing the people whose work creates durable value.
Public Inquiry
Field notes on AI-enhanced technology leadership, governance, institutional change, and what it means to remain humanly accountable as intelligence becomes increasingly capable.
Long before autonomous systems made consequence an architectural question, consequence was literal. I spent two decades in fire, EMS, hazardous-materials response, and rescue service, where preparation, role clarity, communication, trust, and the willingness to act were never abstractions.
I am an ordained Christian minister. My faith informs my obligation to truth, dignity, stewardship, service, and the refusal to treat power as its own justification. It is provenance for my commitments, not a condition of collaboration.
I am a husband, father of nine, and grandfather of six. Legacy is therefore not an abstract ambition. I care about what our systems teach, reward, normalize, and leave behind because people I love—and people I will never meet—will have to live inside their consequences.
I live and work in Southern Maine. The scale of the questions may be civilizational. The test remains intimate: does the system help actual people live, decide, build, heal, and govern more wisely?
If technology now determines enterprise value, institutional risk, strategic viability, or the confidence of the people whose judgment matters, begin with MaineSoftware.
If your concern is how autonomous people, institutions, and agents can act under bounded authority with reviewable evidence, continue into the trust architecture.
If you want to follow the thinking as it develops—leadership, AI, governance, institutional memory, and human becoming—join me through The Diary of @ CTO.
