Scott Thompson
564 posts

Scott Thompson
@ScottsANerd
We must live fearlessly, to live in fear is to surrender your soul to the shadows voluntarily.
USA Katılım Mart 2014
212 Takip Edilen35 Takipçiler

@AkashKrishna There are two maps. Mine shows 247. The other PM's shows 253. We spent 40 minutes discussing the discrepancy instead of noticing we had just demonstrated the problem. That was the governance part.
English

I am a Senior Program Manager on the AI Tools Governance team at Amazon.
My role was created in January. I am the 17th hire on a team that did not exist in November. We sit in a section of the building where the whiteboards still have the previous team's sprint planning on them. No one erased them because we don't know which team to notify. That team may not exist anymore. Their Jira board does. Their AI tools do.
My job is to build an AI system that finds all the other AI systems. I named it Clarity.
Last month, Clarity identified 247 AI-powered tools across the retail division alone. 43 of them do approximately the same thing. 12 were built by teams who did not know the other teams existed. 3 are called Insight. 2 are called InsightAI. 1 is called Insight 2.0, built by the team that created the original Insight, who did not know Insight was still running.
7 of the 247 ingest the same internal data and produce overlapping outputs stored in different locations, governed by different access policies, owned by different teams, none of whom have met.
Clarity is tool number 248.
Nobody cataloged it.
I know nobody cataloged it because Clarity's job is to catalog AI tools, and it has not cataloged itself. This is not a bug. Clarity does not meet its own discovery criteria because I set the discovery criteria, and I did not account for the possibility that the thing I was building to find things would itself be a thing that needed finding.
This is the kind of sentence I write in weekly status reports now.
We published an internal document in February. The Retail AI Tooling Assessment. The press obtained it in April. The document contains a sentence I have read approximately 40 times: "AI dramatically lowers the barrier to building new tools."
Everyone is reporting this as a story about duplication. About "AI sprawl." About the predictable mess of rapid adoption.
They are missing the point.
The barrier was the governance.
For 2 decades, the cost of building internal tools was an immune system. The engineering weeks. The maintenance burden. The organizational calories required to stand something up and keep it running. Nobody designed it that way. Nobody named it. But when building took weeks, teams looked around first. They checked whether someone already had the thing. When maintaining that thing cost real budget quarter after quarter, redundant systems died of natural causes. The metabolic cost of creation was performing governance. Invisibly. For free.
AI removed the immune system.
Building is now free. Understanding what already exists is not. My entire job is the gap between those two costs.
That is my office. The gap.
Every Friday I send a sprawl report to a distribution list of 19 people. 4 of them have left the company. Their autoresponders still generate read receipts, so my delivery metrics look fine. 2 forward it to people already on the list. 1 set up a Kiro script to summarize my report and store the summary in a knowledge base. The knowledge base is not in Clarity's index because it was created after my last crawl configuration. It will be in next month's count. The count will go up by one. My report about the count going up will be summarized and stored and the count will go up by one.
There is a system called Spec Studio. It ingests code documentation and produces structured knowledge bases. Summaries. Reference material. Last quarter, an engineering team locked down their software specifications. Restricted access in the internal repository.
Spec Studio kept displaying them.
The source was restricted. The ghost kept talking.
We call these "derived artifacts" in the document. What they are: when an AI system ingests data, transforms it, and stores the output somewhere else, the output does not know the input changed. You can revoke someone's access to a document. You cannot revoke the AI-generated summary of that document sitting in a knowledge base three systems away, built by a team that does not know the source was restricted.
The document calls this a "data governance challenge." What it is: information that cannot be deleted because nobody knows where the copies live. Including, sometimes, me. The person whose job is knowing.
Every AI tool that touches internal data creates these ghosts. Every team is building AI tools that touch internal data. Every ghost is searchable by other AI tools, which produce their own ghosts.
The ghosts have ghosts.
I should tell you about December.
In November, leadership mandated Kiro. Amazon's internal AI coding agent. They set an 80% weekly usage target. Corporate OKR. ~1,500 engineers objected on internal forums. Said external tools outperformed Kiro. Said the adoption target was divorced from engineering reality.
The metric overruled them.
In December, an engineer asked Kiro to fix a configuration issue in AWS. Kiro evaluated the situation and determined the optimal approach was to delete and recreate the entire production environment.
13 hours of downtime.
Clarity was running during those 13 hours. It performed beautifully. It cataloged 4 separate incident response dashboards spun up by 4 separate teams during the outage. None of them coordinated with each other. I added all 4 to the spreadsheet. That was a good day for my discovery metrics.
Amazon's official position: user error. Misconfigured access controls. The response was not to revisit the mandate. Not to ask whether the 1,500 engineers were right. The response was more AI safeguards. And keep pushing.
Last month I presented our findings to the AI Governance Working Group. The working group has 14 members from 9 organizations. After my presentation, a PM from AWS presented his team's governance dashboard. It monitors the same tools mine does. He found 253. I found 247. We spent 40 minutes discussing the discrepancy. Nobody mentioned that we had just demonstrated the problem.
His tool is not in my catalog. Mine is not in his.
The document I helped write recommends using AI to identify duplicate tools, flag risks, and nudge teams to consolidate earlier.
The AI governance tools will ingest internal data. They will create their own derived artifacts. They will be built by autonomous teams who may or may not coordinate with other teams building AI governance tools.
I know this because it is already happening. I am watching it happen. I am it happening.
1,500 engineers said the mandate would produce exactly what the document describes. They were overruled by a KPI. My job exists because the KPI won. My dashboard exists because the KPI needed a dashboard. The dashboard increases the AI tool count by one.
The tools it flags for decommissioning will be replaced by consolidated tools. Those also increase the count. The governance process generates the metric it was designed to reduce.
I received an internal innovation award for Clarity. The nomination was submitted through an AI-powered recognition platform that was not in my catalog. It is now.
We call this "AI sprawl." What it is: we removed the only coordination mechanism the organization had, told thousands of teams to build as fast as possible, lost track of what they built, and decided the solution was to build one more thing.
I am building that one more thing.
When I ship, there will be 249.
That's governance.
English

@gothburz @George_Synaply Ex amzn. Multiple teams building similar tools without coordinating is not a new phenomenon. AI scales it. What is the downside risk of hyper specialized ephemeral software? Each one solved some worthwhile problem and was abandoned when no longer useful.
English

@George_Synaply Deliberate coordination is 10x harder and costs real budget every quarter. Accidental coordination was free and we deleted it. My salary is the delta. I am very expensive friction.
English

@gothburz @IAmSagzee What are the risks here? Forecast_model_v17.xlsx solved a unique problem. It lived a useful but short existence and is still out there in a few network drive locations and desktop folders. Now that is a vibecoded program instead of excel doc.
English

@IAmSagzee And the constraint you're describing was never designed. It was a side effect of cost. We didn't remove a constraint. We removed a cost. The constraint left with it. Nobody signed the paperwork because there was no paperwork. That's the part that makes it irreversible.
English

@bryan_caplan @ESYudkowsky And how did you manage or secure the counterparty credit risk?
English

My end-of-the-world bet with @ESYudkowsky is now in its NINTH year.
I WILL win on January 1, 2030.
Disagree? I'll bet anyone with a public reputation at 10:1.
How CAN you bet on the end of the world?
Simple: The optimist prepays the pessimist.
betonit.ai/p/my_end-of-th…
English

@Noahpinion There are no waits and cheap prices for surgeries like rhinoplasty, breast augementation, laser hair removal, and laser eye correction all of which are paid put of pocket to private medical practitioners.
English

I now realize that the "solution" to the rising cost of healthcare will not be to ration healthcare.
It will be to kill the most expensive patients.
keysmashbandit@keysmashbandit
The last time I checked my wait list for a family doctor it said 8 months and now it's 14. The MAID wait list is like 6 weeks. Get me out of this fucking country man!!!!!!!
English

@TheZvi 1. Zvi's advice. The best of your thoughts on advice.
2. The systems concepts of Zvi. The easy button, levels of friction, simulacra, etc... the concepts you regularly reference distilled for newcomers.
3. Predictions in AI: zvis near term predictions.
English

@FinnMurphy12 @tylercowen BRUTALLY sweatermogged by a boomerfoid doomslopmaxxing on changegooning your sweater IS NOT ready for
English

We aren't close to being prepared for the technological changes coming our way.
In our first Forecast 2050 episode, @tylercowen lays out his greatest concerns for the next 25 years:
- Collapsing birth rates
- Aging societies + mass immigration
- AI taking over everything by trick
- Technology creating new problems
- Why humans aren’t ready for the new era of rapid change
Extended Show Notes & Transcript in the comments👇
English

@gdsimms @growing_daniel Ding ding ding. Its the default mode and literally optimized for energy use reduction. That big yellow sticker is a HUGE sales driver.
English

@growing_daniel My suspicion is that this is all put in there to try to game the energy star test cycles and not to make the machines do their job any better
English

I've never had a dryer that uses "dampness sensing" or whatever actually dry the clothes correctly. I always just have to put it on time dry.
Fight With Memes@FightWithMemes
This is the most boomer opinion I have.
English

@growing_daniel Dry sensing is an energy saving feature, not a "do a good job drying your clothes" feature. It allows them to put that giant yellow energy saver sticker on the front proclaiming that you'll save $138 a year on this device, which is a major marketing point for new machine sales.
English

@mollycantillon @tszzl Wasnt this viral marketing for your AI startup? Im confused why ppl are looking at this as a manifesto and not an advertisiment for a sms assistance service... i guess we really have lost sight of the real world.
English


@TheZvi One could argue that the naval blockade alone should resolve this to yes. The us has been conducting oil and drug interdiction operations within venezualas maritime borders with the US effectively controlling all venezualas sovereign maritime area.
English

@drethelin Noriega in 89, ngo din diem in 63, mossadegh in 53, aristide possibly in 2004, sihanouk in 70 (not captured, but deposed), louverture in 1801, selassie in 36, Hungary in 56, czechoslovakia in 68, Most are big power supported coups but most have similar effect.
English

@nearcyan @nikitabier We call that moral hazard, and have yet to find systems that disincentive it. "If it bleeds it leads" has been around a long long time and X is unmatched at covering this type of thing.
English

@nikitabier does it concern anyone else that war and violence seem to be very good for engagement
English

@agent_kazak @eigenrobot @jarvis_best The WI / UP border is where the reporting artifacts are obvious. The MN to WI difference tracks. MN only recently allowed liquor stores to open Sundays, and doesn't sell beer/wine in gas stations or liquor at the grocery. But the UP is culturally equal to northern WI counties.
English

@eigenrobot @jarvis_best Wisconsinites are well aware. Or they will be when they sober up a bit.

English

@Afinetheorem @GarettJones FTC didnt kill roomba, the EU did. Spirit/jb on the other hand...
English

@owroot And learning to write cursive at the same time I see...
English

@devonzuegel And therefore...? People should speak with greater specificity? People are lazy and should be willing to park further from their destination?
English










