Category Archives: Sailing

Let AI Dangle: Why the sketch.dev Integrity Breach Demands Human Accountability, Not Technical Cages

AI safety should not be framed as choosing between safety and capability when it’s more accurately between the false security of constrained tools and the true security of accountable humans using powerful tools wisely. We know which choice builds better software and better organizations. History tells us who wins and why. The question is whether we have the courage to choose freedom of democratic systems over the comfortable illusion of a fascist control fetish.

“Let him have it” Chris – those few words destroyed a young man’s life in 1952 because their meaning was fatally ambiguous, as famously memorialized by Elvis Costello in his hit song “Let Him Dangle”.

Did Derek Bentley tell his friend to surrender the gun or to shoot the police officer? The dangerous ambiguity of language is what led to a tragic miscarriage of justice.

Today, we face a familiar crisis of contextualized intelligence, but this time it’s not human code that’s ambiguous, it’s the derived machine code. The recent sketch.dev outage, caused by an LLM switching “break” to “continue” during code refactor, represents something far more serious than a simple bug.

This is a small enough change in a larger code movement that we didn’t notice it during code review.

We as an industry could use better tooling on this front. Git will detect move-and-change at the file level, but not at the patch hunk level, even for pretty large hunks. (To be fair, there are API challenges.)

It’s very easy to miss important changes in a sea of green and red that’s otherwise mostly identical. That’s why we have diffs in the first place.

This kind of error has bitten me before, far before LLMs were around. But this problem is exacerbated by LLM coding agents. A human doing this refactor would select the original text, cut it, move to the new file, and paste it. Any changes after that would be intentional.

LLM coding agents work by writing patches. That means that to move code, they write two patches, a deletion and an insertion. This leaves room for transcription errors.

This is another glaring example of an old category of systemic failure that has been mostly ignored, at least outside nation-state intelligence operations: integrity breaches.

The real problem isn’t the AI because it’s the commercial sector’s abandonment of human accountability in development processes.

The common person’s bad intelligence is a luxury that is evaporating rapidly in the market. The debt of ignorance is rising rapidly due to automation.

The False Security of Technical Controls

When sketch.dev’s team responded to their AI-induced outage by adding “clipboard support to force byte-for-byte copying,” they made the classic mistake of treating a human process problem with a short-sighted technical band-aid. Imagine if the NSA reacted to a signals gathering failure by moving agents into your house.

The Stasi at work in a mobile observation unit. Source: DW. “BArch, MfS, HA II, Nr. 40000, S. 20, Bild 2”

This is like responding to a car accident by lowering all speed limits to 5 mph. Yes, certain risks can be reduced by heavily taxing all movements, but it also defeats the entire purpose of having movement highly automated.

As the battle-weary Eisenhower, who called for “confederation of mutual trust and respect”, also warned us:

If you want total security, go to prison. There you’re fed, clothed, given medical care and so on. The only thing lacking… is freedom.

Constraining AI to byte-perfect transcription isn’t security. It’s not, it really isn’t. It’s surrendering the very capabilities that make AI valuable in the first place, lowering security and productivity with a loss-loss outcome.

My father always used to tell me “a ship is safe in harbor, but that’s not what ships are built for”. When I sailed across the Pacific, every day a survival lesson, I knew exactly what he meant. We build AI coding tools to intelligently navigate the vast ocean of software complexity, not to sit safely docked at the pier in our pressed pink shorts partying to the saccharin yacht rock of find-and-replace operations.

Turkey Red and Madder dyes were used for uniforms, from railway coveralls to navy and military gear, as a low-cost method to obscure evidence of hard labor. New England elites (“Nantucket Reds”) ironically adapted them to be a carefully cultivated symbol of power. The practical application in hard labor inverted to a subtle marker of largess, American racism of a privileged caste.

The Accountability Vacuum

The real issue revealed by the sketch.dev incident isn’t that the AI made an interpretation – it’s that no human took responsibility for that interpretation.

The code was reviewed by a human, merged by a human, and deployed by a human. At each step, there was an opportunity for someone to own the decision and catch the error.

Instead, we’re creating systems where humans abdicate responsibility to AI, then blame the AI when things go wrong.

This is unethical and exactly backwards.

Consider what actually happened:

  • AI made a reasonable interpretation of ambiguous intent
  • A human reviewer glanced at a large diff and missed a critical change
  • The deployment process treated AI-generated code as equivalent to human-written code
  • When problems arose, the response was to constrain the AI rather than improve human oversight

The Pattern We Should Recognize

Privacy breaches follow predictable patterns not because systems lack technical controls, but because organizations lack accountability structures. A firewall that doesn’t “deny all” by default isn’t a technical failure, because we know all too well (e.g. codified in privacy breach laws) it’s organizational failure. Someone made the decision to configure it that way, and someone else failed to audit that very human decision.

The same is true for AI integrity breaches. They’re not inevitable technical failures because they’re predictable organizational failures. When we treat AI output as detached magic that humans can’t be expected to understand or verify, we create exactly the conditions for catastrophic mistakes.

Remember the phrase guns don’t kill people?

The Intelligence Partnership Model

The solution isn’t to lobotomize our AI tools into ASS (Artificially Stupid Systems) it’s to establish clear accountability for their use. This means:

Human ownership of AI decisions: Every AI-generated code change should have a named human who vouches for its correctness and takes responsibility for its consequences.

Graduated trust models: AI suggestions for trivial changes (formatting, variable renaming) can have lighter review than AI suggestions for logic changes (control flow, error handling).

Explicit verification requirements: Critical code paths should require human verification of AI changes, not just human approval of diffs.

Learning from errors: When AI makes mistakes, the focus should be on improving human oversight processes, not constraining AI capabilities.

Clear escalation paths: When humans don’t understand what AI is doing, there should be clear processes for getting help or rejecting the change entirely.

And none of this is novel, or innovative. This comes from a century of state-run intelligence operations within democratic societies winning wars against fascism. Study the history of disinformation and deception in warfare long enough and you’re condemned to see the mistakes being repeated today.

The Table Stakes

Here’s what’s really at stake: If we respond to AI integrity breaches by constraining AI systems to simple, “safe” operations, we’ll lose the transformative potential of AI-assisted development. We’ll end up with expensive autocomplete tools instead of genuine coding partners.

But if we maintain AI capabilities while building proper accountability structures, we can have both safety and progress. The sketch.dev team should have responded by improving their code review process, not by constraining their AI to byte-perfect copying.

Let Them Have Freedom

Derek Bentley died because the legal system failed to account for human responsibility in ambiguous situations. The judge, jury, and Home Secretary all had opportunities to recognize the ambiguity and choose mercy over rigid application of rules. Instead, they abdicated moral responsibility to legal mechanism.

We’re making the same mistake with AI systems. When an AI makes an ambiguous interpretation, the answer isn’t to eliminate ambiguity through technical constraints when it’s to ensure humans take responsibility for resolving that ambiguity appropriately.

The phrase “let him have it” was dangerous because it placed a life-or-death decision in the hands of someone without proper judgment or accountability. Today, we’re placing system-critical decisions in the hands of AI without proper human judgment or accountability.

We shouldn’t accept the kind of world where we eliminate ambiguity, as if a world without art could even exist, so let’s ensure someone competent and accountable can be authorized to interpret it correctly.

Real Security of Ike

True security comes from having humans who understand their tools, take ownership of their decisions, and learn from their mistakes. It doesn’t come from building technical cages that prevent those tools from being useful.

AI integrity breaches will continue until we accept that the problem is humans who abdicate their responsibility to understand and verify what is happening under their authority. The sketch.dev incident should be a wake-up call for better human processes, more ethics, not an excuse for replacing legs with pegs.

A ship may be safe in harbor, but we build ships to sail. Let’s build AI systems that can navigate the complexity of real software development, and let’s build human processes to navigate the complexity of working with those systems responsibly… like it’s 1925 again.

North Korea Says Capsized Russian-Designed Destroyer Was Only “Scratched”

You can’t make this Trump-sounding tinpot dictator stuff up. The second big new North Korean destroyer, designed by Russian spies stealing American ideas, was launched sideways and damaged, capsized, and has been taking on water.

KCNA reported that there were no holes on the ship’s bottom – contrary to initial reports.

“The hull starboard was scratched and a certain amount of seawater flowed into the stern section,” the agency said.

Seawater entered the hull but there was no hole? It’s like a ship taking on seawater through a… not hole. Nobody is allowed to say there’s a hole, when water flows into the stern, so that means ’tis but a scratch.

The Black Knight sketch was supposed to be fictional humor.

Arthur delivers a mighty blow which completely severs the Black Knight’s left arm at the shoulder. Arthur steps back triumphantly.

Arthur: “Now stand aside worthy adversary.”

Black Knight (Glancing at his shoulder): “Tis but a scratch.

Arthur: “A scratch? Your arm’s off.”

Black Knight: “No, it isn’t.”

Arthur (Pointing to the arm on ground): “Well, what’s that then?”

Black Knight: “I’ve had worse.”

Arthur: “You’re a liar.”

Black Knight: “Come on you pansy!”

“‘Tis but a scratch…. Just a flesh wound.” Source: Black Knight, Monty Python

I hate to ask out loud, but am I the only one seeing the 1982 Russian integrity vulnerability, when looking at that capsized hulk?

In time the Soviets came to understand that they had been stealing bogus technology, but now what were they to do? By implication, every cell of the Soviet leviathan might be infected. They had no way of knowing which equipment was sound, which was bogus. All was suspect, which was the intended endgame for the entire operation.

I suspect North Korea now will send even more of their boys to die in a ditch in Ukraine, until Russia gives them a new bogus stolen ship design. “This time with launch instructions” someone must be screaming in Russian right about now.

Drone Delivers Life Preserver to Drowning Swimmer

There’s a good case to be made that life preservers on a public beach could be attached to a delivery drone as a standard procedure now.

Smith noticed several emergency life preservers placed along the beach. Acting quickly, he grabbed one and used his drone to fly it out over the water. His first attempt missed the girl, but on the second try, he was able to lower the preserver within her reach.

The story says the drone operator just wants to go back to killing sharks, but really he should start a company focused on saving humans. Imagine boats having emergency drones too, including breathing apparatus.

Woefully Inadequate Intelligence Analysis in “Worse Than Signalgate”

Ryback’s “Worse Than Signalgate” analysis in The Atlantic of the Zimmermann Telegram affair represents the kind of simplistic, great-man historiography that we’ve spent decades trying to eradicate from serious historical discourse.

The article’s framing of the incident as mere diplomatic bungling rather than a calculated intelligence operation is not merely incomplete—it’s intellectually dishonest.

Let me be perfectly clear: The Zimmermann Telegram was not primarily a story of German incompetence, but rather of British intelligence brilliance in a targeted campaign to neutralize pro-German sympathies in America, particularly those held by influential figures like President Wilson and industrialist Henry Ford.

For those unfamiliar with this pivotal event: In January 1917, German Foreign Minister Arthur Zimmermann sent an encrypted message proposing that Mexico join Germany against the United States if America entered World War I. Germany promised to help Mexico recover territories lost to the U.S. (Texas, New Mexico, and Arizona) and suggested Mexico invite Japan to join this alliance. British intelligence intercepted and decoded this message, then strategically released it to the American public through the press in February 1917. The telegram’s content was so incendiary that it helped overcome Wilson’s resistance to war, despite his previous stance against intervention. What Ryback fails to grasp is that this was not simply a German diplomatic blunder, but a calculated British intelligence operation designed to circumvent pro-German American leadership.

Nigel de Grey, in Room 40 Old Building, was one of the first British officers to partially decrypt the Zimmermann Telegram. William Montgomery then completed the decryption and verified its authenticity, which led Admiral Sir Reginald “Blinker” Hall to devise a cunning plan for public exposure.

It would be like today decoding a Putin memo that revealed Russia interfered with the Presidential election to put Trump in office, is behind the NRA pushing guns even as mass murders spread, and funds campaigns to get Texas and California to secede. Or perhaps even more shrewdly, points out the heated campaign to transfer corporations and manufacturing to Texas is foreshadowing of a Russian campaign for Texas to secede like Brexit.

Ryback’s article conveniently omits there was an extensive German sabotage campaign on American soil—over 40 documented bombing incidents between 1914 and 1917, including the Black Tom explosion that damaged the Statue of Liberty. Wilson’s administration consistently intentionally misdirected blame toward leftists, anarchists, and labor agitators while knowing full well that German agents were the actual perpetrators.

The July 1916 Preparedness Day bombing in San Francisco serves as a perfect example of this deliberate misdirection, with local labor leaders Thomas Mooney and Warren Billings falsely imprisoned while clear ties to German involvement was suppressed.

What makes Wilson’s denialism all the more egregious is that ordinary Americans were acutely aware of the wave of suspicious fires, explosions, and acts of sabotage sweeping the nation. Newspapers from coast to coast reported on these “accidents” at munitions plants, shipping facilities, and infrastructure targets. The pattern was unmistakable to anyone paying attention – a veritable domestic Holocaust of German-orchestrated destruction.

While Americans buried their dead following the Black Tom explosion and the Kingsland munitions plant fire, Wilson’s administration continued its calculated misdirection campaign, attributing these acts to labor unrest or simple industrial accidents. This deliberate obfuscation occurred even as federal investigators privately compiled evidence of German involvement.

When British intelligence finally bypassed Wilson with the Zimmermann Telegram revelation, it wasn’t just exposing German intentions—it was circumventing an American president who had systematically lied to his own citizenry about clear and present dangers on American soil.

Ford’s pro-German involvement is particularly egregious in its omission. Far from being a mere “isolationist,” Ford was actively operating as anti-American, accepting millions in German funds while failing to deliver promised agricultural equipment, as harshly pointed out by congressional rebukes.

His anti-Semitic publications aligned with and promoted German propaganda aims (Hitler and Goebbels both credited Ford for influencing them). His opposition to American entry into the war wasn’t simply isolationism or pacifism but part of a broader ideological position that actively worked to undermine American principles and security interests.

Wilson himself had campaigned on keeping America out of war, implicitly to aid German militant aims, and maintained back-channel communications with German officials well into 1917.

The British intelligence operation around the Zimmermann Telegram therefore must be understood in full context as a sophisticated psychological operation targeting the American public directly, deliberately circumventing high profile pro-German sympathizers even ones in positions of oligarchial power.

Room 40’s work wasn’t merely clever codebreaking; it was a masterful influence operation that recognized the power of public opinion over elite preferences in American politics.

A simple illustration of German military plans destroyed Henry Ford and Woodrow Wilson attempts to keep America aligned with Germany.

Ryback’s framing of Zimmermann as merely delusional misses the obvious wider German strategy of global destabilization—from their support of Irish republicans to their activation of terror networks in British India. The telegram was but one component within a sophisticated global strategy, far from incompetent, that the British correctly identified and countered.

This breezy Atlantic article offers a superficial, decontextualized history that contributes unnecessary flap and noise to a proper understanding of highly skilled intelligence operations, directed public opinion manipulation, or the genuine complexity of Ford and Wilson’s toxic false neutrality. They weren’t neutral, they were actively and directly harming Americans.

The incongruity in Wilson’s leadership – reluctance to confront foreign threats while eagerly suppressing domestic groups – is the cruel understated aspect of his presidency that complicates any conventional narrative of Wilson as principled.

His hesitation to directly condemn German actions, even after American ships were sunk, contrasts sharply with his administration’s swift and brutal responses to perceived domestic threats.

Wilson’s administration was remarkably aggressive in suppressing leftist political movements. The Espionage Act of 1917 and Sedition Act of 1918 were used to imprison hundreds of labor activists, socialists, and anti-war protesters. Eugene V. Debs, who had received nearly a million votes as the Socialist Party presidential candidate, was sentenced to 10 years in prison essentially for giving an anti-war speech.

And yes, Wilson’s racial policies were particularly devastating. His administration resegregated federal offices that had been integrated since Reconstruction. He screened “Birth of a Nation” at the White House, effectively endorsing its racist narrative and helping fuel the revival of the KKK.

The 1919 Elaine Massacre is a perfect example. When Black sharecroppers in Arkansas organized to demand fair payment for their cotton, white mobs attacked them. Rather than protecting these American citizens being attacked for exercising their rights, the Army was deployed by Wilson to attack them further, resulting in the shooting deaths of hundreds of Black Americans by federal troops.

This pattern reveals a deeply troubling aspect of Wilson’s worldview when the Zimmerman telegram landed. The President seemed far more willing to extend understanding and restraint to certain powers, even hostile ones, than to extend basic constitutional protections to Black Americans or his political opponents at home.


Technical Appendix: The British Room 40 Operation in Detail


I’ve called the British intelligence operation clever, but let me elaborate on why this characterization is warranted. While a Zimmermann Telegram brought the power of the public in the United States to bear down on the deeply racist and corrupted President Wilson and his cronies, the British had managed this all without blowing their operation. Room 40 didn’t just see the message and decode it, they used simple espionage theater to keep their war-winning methods a secret.

  • Double Intercept: both an original high-strength cracked cipher version and a weakened known-vulnerable copy from Mexico were used. The latter “internal” communication between America and Mexico gave the British a cover story for how and where they “found” it.
  • Deception by Omission: The Americans were told a true story about American officials caught working on the side of Germany (which meant against America), yet were not told that German communications had been cracked.
  • Protected Assets: Keeping Room 40 a secret meant the British preserved a massive intelligence edge for the rest of the war, similar to how Polish intelligence cracked the Nazi Enigma in WWII and yet are rarely if ever credited properly even to this day.
  • Let the Enemy Fall Into It: The telegram’s authenticity was confirmed by Zimmermann, unaware of British methods.

Although the codebreaking unit in Room 40 had cracked and validated Germany’s top diplomatic cipher, they couldn’t let the Germans find out and shut down a critical intelligence stream. The telegram from Germany to Mexico proposing an alliance against the U.S. needed another path as a plausible origin story.

  • August 5, 1914 — British cable ship HMTS Alert was ordered by Admiralty to cut Germany’s five undersea telegraph cables, just one day after declaration of war. These were the major transatlantic cables that connected Germany directly to North America and other global regions. Notably, Germany saw America secretly as an ally against Britain and thus switched communications to its “neutral” lines, which meant even using American diplomatic channels on the American cables. That’s exactly where British intelligence was listening.
  • January 16, 1917 — Zimmermann’s telegram routes from Berlin to the German ambassador in Washington DC, via the U.S. diplomatic cables that it believed to be secure from the British because American. British intelligence intercept this message and decipher it, realizing important value.
  • February 1–3, 1917 — British agents quietly obtain a second version of the telegram that had been forwarded from Washington DC to Mexico City, due to a less secure cipher widely known to be compromised. Germany also angrily announces that it will immediately resume unrestricted submarine warfare to sink passenger ships, prompting President Wilson to declare to the Senate that he is totally surprised by such a thing and waiting to see any evidence of “actual overt acts” of harm.

    I cannot bring myself to believe that [Germany] will indeed pay no regard to the ancient friendship between their people and our own or to the solemn obligations which have been exchanged between them, and destroy American ships, and take the lives of American citizens in the wilful prosecution of the ruthless naval program they have announced their intention to adopt. Only actual overt acts on their part can make me believe it even now.

    Immediately, German U-boats attacked and sank the Housatonic, and then three days later attacked and sank the 470-ft passenger ship SS California off the Irish coast, killing nearly 50 people in just 10 minutes. Yet again, for at least the second year, Wilson’s calculated and cruel inaction leaves American blood on his hands. On the 12th the schooner Lyman M. Law is sunk.

  • February 19, 1917 — Britain’s Admiral Hall presents the Zimmerman telegram to U.S. officials, explaining it had been discovered in Mexico coming from American diplomatic channels.
  • March 1, 1917 — Unable to continue pretending it has a “neutral” role while allowing years of German attacks killing Americans, alarms about the telegram are rising, and the British propel the news all the way into U.S. newspaper headlines.
  • March 3, 1917 — Zimmermann inexplicably confirms the telegram’s authenticity, removing all doubt in the public eye.
  • March 12, 1917 — steamship Algonquin is sunk, followed four days later by the sinking of the US steamship Vigilancia without warning, killing 15 (including six Americans), and soon after the sinking of the US oil tanker Healdton, killing 21 (including seven Americans).
  • April 6, 1917 — unable to waffle and wiggle any longer, having knowingly allowed horrible attacks on Americans to run unanswered for years, Wilson finally is compelled by public outrage to declare war on Germany.

The Atlantic article fundamentally misses that the Zimmermann event represented a meticulously planned intelligence operation by Britain’s top talent, who embedded a public sentiment campaign within a seemingly neutral document while secretly preserving their ability to continue decoding German communications.

This wasn’t simply German incompetence meeting British luck—it was strategic calculation meeting strategic counter-calculation, a far more nuanced and historically significant interaction. In fact, the deep and careful British intelligence work was so effective that, even a century later, some historians still appear to be confused. They portray Zimmermann as incompetent rather than someone with reasonable expectations of ongoing American sympathy through the toxicity of an extremist pro-German President Wilson—even for Germany’s own territorial ambitions (a historical parallel with differently disturbing modern echoes, like how Trump serves Russia today).