Insights · Why AI built for work
ChatGPT is brilliant. So why can't it do your company's work?
Ask a general-purpose AI service, "What were sales at each of our branches this month?" and the answer comes back in seconds.
Blog
Insights
Short reads for decision makers bringing AI into the business. Each starts from a question teams actually run into and shows, in three minutes, what gets solved and what is different.
Part 1
Why AI built for work
How AI built for work differs from general-purpose AI, and the questions that frame the whole series.
Insights · Why AI built for work
Inventory right now: does the AI chatbot know today's count?
A sales rep asks the company's AI chatbot, "How many units of product A do we have in stock right now?" The ch…
Insights · Why AI built for work
Data scattered across systems: can AI gather it in one pass?
"Pull the accounts more than 90 days overdue this quarter that our policy says go to collections, and group th…
Part 2
Asking your company database
Connecting company databases safely, asking in plain language, and getting accurate answers: internal and Amazon RDS or Supabase databases, accurate SQL, department-specific definitions, and what security teams ask before a production database is connected.
Insights · Asking your company database
The AI's SQL runs fine. Why is the revenue number wrong?
The demo is impressive.
Insights · Asking your company database
Departments count revenue differently. Whose number will the AI use?
At the executive meeting, the sales division reports that revenue grew 12%, and the finance team says 8%.
Insights · Asking your company database
Put AI on the production database, and security asks three questions
When someone proposes connecting AI to a business database, the DBAs and the security team raise the same three questions.
Insights · Asking your company database
How do you connect your company database securely?
In a meeting to review an AI rollout, the IT team gets a request: "Open the database port so the AI server can…
Insights · Asking your company database
Connecting Amazon RDS to OwlAI, the easy way
Your customer database runs on Amazon RDS.
Insights · Asking your company database
Connecting Supabase to OwlAI, the easy way
Your product data lives in Supabase.
Part 3
Better with OwlDB
OwlAI alone can connect to company databases and answer questions. Adding OwlDB, a database management solution, brings table-level permissions, protection from heavy queries, standby node use, and connections without credentials. This part covers what improves and how.
Insights · Better with OwlDB
Open a database to one team. Do the off-limits tables come along?
An administrator opens the business database to the sales role.
Insights · Better with OwlDB
Could one AI query slow down your order screen?
When the AI starts scanning a table of hundreds of millions of rows on your production database with no filter…
Insights · Better with OwlDB
The standby database: can it take AI queries off production?
Ask for "sales by region last month," and the AI doesn't pull the answer right away.
Insights · Better with OwlDB
The database address changes: does the AI connection follow?
Over the weekend, the database server moved to new hardware.
Part 4
AI that improves with use
How individual and team knowledge accumulates so answers get better, and how the company keeps control of that memory.
Insights · AI that improves with use
Report preferences you repeat every time: can AI remember them?
Each time you ask an AI for a report draft, you attach the same conditions: "No more than three tables, conclu…
Insights · AI that improves with use
Only the old hand knew the data rules. Will the AI still know them?
Every company has a veteran who knows the traps in its data.
Insights · AI that improves with use
Criteria agreed in an earlier chat: can AI pick them up again?
Last week you spent a long conversation with the AI working out how to look up receivables by customer.
Insights · AI that improves with use
The account is locked. Does the AI still remember what they told it?
A sales rep has left your company.
Part 5
Research and delegating work
From researching documents and the web with sources attached, to handing off real work such as recurring reports and documents in company templates.
Insights · Research and delegating work
Wide-ranging research: can AI cover it without missing parts?
"Summarize e-payment regulation trends in three Southeast Asian countries and where the major competitors stan…
Insights · Research and delegating work
Numbers for a report: can you trust the sources AI cites?
You're about to paste market data the AI summarized into a report when your manager asks, "Where's this number…
Insights · Research and delegating work
Could AI take over the weekly report end to end?
Every Monday morning, someone on your team does the same job.
Insights · Research and delegating work
The AI wrote the document. Why paste it into the company template?
Ask an AI for this month's quote and the content arrives fast.
Part 6
Everyday conversations and meetings
Trust problems that come up in daily use, such as long conversations, instructions added mid-conversation, and wrong answers that are easy to miss, plus the path from meeting recordings to minutes.
Insights · Everyday conversations and meetings
Why does the AI forget your opening conditions as the chat grows long?
You have spent an hour going through quarterly results with an AI.
Insights · Everyday conversations and meetings
Remembered one more condition? Must the AI answer start over?
You send "Put last quarter's sales by region into Excel," and the moment it's gone you remember: online sales should have been left out.
Insights · Everyday conversations and meetings
Tens of thousands of orders: does AI really read them all?
You ask an AI to look through this year's orders and find anything unusual.
Insights · Everyday conversations and meetings
Minutes keep getting in-house terms wrong. Must you fix them by hand?
A meeting ends and you open the minutes that were written automatically.
Insights · Everyday conversations and meetings
Could AI turn a meeting transcript into usable minutes?
You recorded an hour-long meeting and got it transcribed. What you have now is dozens of pages of text.
Part 7
Security and audit
What is blocked by rules, what is left to AI judgment, and how records confirm it: data egress, hidden instructions, personal data, passwords and tokens, sandboxed code execution, approvals, external connections, and audit records.
Insights · Security and audit
AI is sometimes wrong. Can it handle work involving customer data?
AI evaluations for work usually end on the same question: the performance is clear, but can we really trust it with this?
Insights · Security and audit
You want the newest AI models. The customer data can't leave.
A sales rep pastes a customer list into an AI and asks it to tidy things up.
Insights · Security and audit
A hidden line in an email you asked it to summarize. Will the AI obey?
You ask the AI to summarize twenty emails from business partners.
Insights · Security and audit
Masked in chat, do national ID numbers still turn up in Excel?
You ask the AI to analyze customer inquiry data, and its answer prints customer phone numbers verbatim.
Insights · Security and audit
Do the passwords and keys you give the AI stay in the chat logs?
Connect the AI to your company database and a cloud drive, and you have handed it the database password and access to the drive.
Insights · Security and audit
The AI runs its own code to tidy your Excel. Is it safe unreviewed?
Upload a sales spreadsheet and ask for regional growth rates on a single PowerPoint slide, and the AI can't answer in words.
Insights · Security and audit
Email can't be unsent. So can you let an AI hit send?
Say you ask the AI to "let everyone from last week's meeting know the schedule changed," and it misreads the a…
Insights · Security and audit
Once the AI sends email and edits tickets, who approves it?
"File last week's meeting action items as issues." "Find open slots in the team's calendar next week." Once em…
Insights · Security and audit
Asking AI for data you aren't allowed to see: what happens?
A sales rep who has no access to HR data asks the AI, "What's the average salary increase this year?" If the A…
Part 8
Adoption and operations
From choosing a deployment model to installation, rollout, updates, and a reference case.
Insights · Adoption and operations
Security says no to the cloud. Where does your business AI go?
When a company evaluates business AI, the discussion tends to drag once someone asks, "Can we actually put thi…
Insights · Adoption and operations
Bringing AI to a network where ChatGPT won't load
Type the ChatGPT address on a PC on your business network, and the page doesn't load.
Insights · Adoption and operations
While IT and security wait on each other, the AI rollout drags on
At the first meeting after your company decides to bring in business AI, IT asks which server to prepare, secu…
Insights · Adoption and operations
Updating self-hosted AI: can it be done safely?
The notice says a new version is out, yet the server is still sitting several versions back.
Insights · Adoption and operations
What changed when 163 employees asked AI 3,541 questions a month
On an average weekday, 152 questions reach the company's internal OwlAI.
Insights · Adoption and operations
Every business AI looks alike in a demo. Ten questions to ask first
Ask any product a short question in a demo, such as "Who were our top customers by revenue last month?", and you get a plausible answer.