To fix your company's AI problems, start at the top.
To fix your company's AI problems, start at the top.
The latest issue of Complex Machinery: The bot doesn't always have to talk
You didn't build it yourself, so you have to take some things on faith.
To borrow a phrase about the cloud ...
The latest issue of Complex Machinery: AI is magic. In various meanings of the word.
The latest issue of Complex Machinery: Sometimes AI is a better fit for that task. Emphasis on "sometimes"
The latest issue of Complex Machinery: Machines conquered Wall Street, then learned to play a mean game of poker.
The latest issue of Complex Machinery: AI use cases, as driven by fraud and stunts. And squeeze bottles.
The latest issue of Complex Machinery: Cookies can tell us a lot about AI use cases. Especially the more questionable variety.
The latest issue of Complex Machinery: More lessons from the CrowdStrike incident
This is a good time to move at your own pace
Wait, two _trillion_ dollars??
In AI, as elsewhere, execution is what matters
The latest issue of Complex Machinery: How the CrowdStrike incident reflects a key risk in complex systems
The latest issue of Complex Machinery: The shady side of the AI business.
The latest issue of Complex Machinery: Understanding the race against time that is AI's popularity.
The latest issue of Complex Machinery: Pondering the easy money in AI, and when it will come due.
The latest issue of Complex Machinery: Google changes path on its AI Overviews product
The latest issue of Complex Machinery: the new GPT-4o trips, Google's AI Overview stumbles, and asking what-if questions when building AI products
Is AI having its bubble moment? If so, why would that matter?
The latest issue of Complex Machinery: Some short-term AI annoyances, plus some optimism about the future
The latest issue of Complex Machinery: Pricing out the risk of your latest AI adventure
The latest issue of Complex Machinery: A look into auto manufacturers passing driver details to data brokers
The latest issue of Complex Machinery: thinking about our approaching bot-on-bot future
The latest issue of Complex Machinery: the randomness inside every AI model
The latest issue of Complex Machinery: deepfakes for crime, facial recognition, and minding the robots
Complex Machinery is my new newsletter on AI, risk, and related topics.
AI chatbots are great, but they're still a little rough around the edges
Let's remember the human element in AI-driven content moderation
Wider AI truths, as surfaced by LLM failures
What the stock photo company's new service tells us about safe-to-use datasets
What's the next step in AI?
A handful of all-too-common pitfalls in hiring.
Releasing an AI chatbot exposes your company to new risks. Here are some ideas on how to handle them.
If you want the most out of AI, you need to be strategic about how you employ it.
A reminder of generative AI's chaotic potential
The artist currently and formerly known as "AI" gets another turn on the stage.
What custom software can tell us about bringing AI into a company.
My take on tools such as Dall-E, Stable Diffusion, and ChatGPT
Risk mitigation for your ML/AI projects
Automated machine learning (autoML) and its impact on the ML/AI space
A short list of ways an ML/AI modeling project can go off the rails
Is it time to update your ML models?
Why this question deserves a deeper conversation.
Using spreadsheets to explain core ML/AI concepts to executives.
Some datasets are problems in waiting.
Potential problems that can affect the entire field.
Sometimes, you have to tackle a challenge head-on.
Keeping the bigger picture in mind.
My panel discussion with Linda Liu (Hyrecar) and Giacomo Vianello (Cape Analytics)
Zillow recently closed down its home-buying program. Do I see this as a failure of ML/AI? In a word: No.
I've published an article on O'Reilly Radar: The key ingredient to a successful remote team? Leadership buy-in.
I interviewed product manager Chris Butler about the role uncertainty plays in AI product management.
Best practices to balance the risk and reward of building predictive models.
Spotting opportunities to build AI systems that complement, not outright replace, people on the job.
BI is periods. AI is question marks. Simulation is ellipses.
When talking with your company's data scientists, does the conversation quickly bog down? Try these questions to keep things moving.
Want to improve your risk assessment? Identify, then question, the constants in your world.
Nervous about meeting with your company's execs, legal counsel, or product team? Just answer these four questions.
I've published an article on O'Reilly Radar: how many times will we rename the data field?
Three steps can reduce churn in your company's data leadership role(s).
Humans versus machines? To reduce your risk, the best answer is "yes."
As we learn more about AI, what will change about how we develop and deploy it?
In search of ML/AI success? Know your hard and your soft numbers.
Does the AI hype meet the technical term of a "bubble?"
Putting AI to good use in a dangerous environment.
Lemonade's recent media spotlight is a cautionary tale for any company using ML/AI.
Discipline pays off.
There's a lot more to this than just building models.
A little planning will go a long way.
Go off the beaten path to make the most of your data-related hiring.
Is your company getting started with ML/AI? These uncommon tips will save you time and trouble.
It's not just train-test-deploy
Why it's important to be able to simulate your own data.
Determine whether an ML/AI project is for automation or for innovation, so you can prioritize it accordingly.
Punctuation matters when working with data: BI is periods. AI is question marks.
Understand which aspects of your ML/AI shop can (and cannot) give you an edge over the competition.
If your company has several ML/AI efforts on the roadmap, it can be difficult to decide how to prioritize them. You can look to the stock market for guidance.
It's bad enough when the model is wrong; even worse, when it's wrong and you didn't have to build it in the first place.
Explaining the realities of how an ML/AI project may go awry.
Shedding light on the hidden costs of employing ML/AI models, which can upend the price/value ratio.
When a group predicts students' performance in lieu of holding a exam, it leads to some lessons in the deployment and use of ML/AI models.
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 9 in a series)
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 8 in a series)
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 7 in a series)
ML/AI models still require a lot of human involvement.
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 6 in a series)
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 5 in a series)
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 4 in a series)
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 3 in a series)
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 2 in a series.)
What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 1 in a series)
Three questions to help you gauge your risk of being laid off.
Business models involving ML/AI are sitting on unrecognized risk.
COVID-19 may be what gets companies to take their ML/AI efforts seriously.
The AI world is sitting on all kinds of risk ... but no one wants to talk about it.
Should your company's data scientists hold a PhD? Probably not.
Executives want to know how to employ ML/AI in their company. They need more than just quick tips.
Since "more data is better," what do I do if I don't have enough?
How much data do you need to build good predictive models?
Looking for a data science job? It involves far more than the technical know-how.
Looking at data ethics through the lens of risk. (Part 5 in a series.)
Looking at data ethics through the lens of risk. (Part 4 in a series.)
Looking at data ethics through the lens of risk. (Part 3 in a series.)
Looking at data ethics through the lens of risk. (Part 2 in a series.)
Looking at data ethics through the lens of risk. (Part 1 in a series.)
How can business intelligence (BI) launch your data efforts, and pave the way for your first data science hire?
Having trouble hiring data scientists? or, once you hire them, do they not stick around? You may be tripping over your own feet. Part 2 of 2.
Having trouble hiring data scientists? or, once you hire them, do they not stick around? You may be tripping over your own feet. Part 1 of 2.
Having trouble hiring data scientists? Borrow some ideas from your sales team.
A successful data science shop requires more than just data scientists.
The what, why, and how of a data strategy -- a road map for your company's data efforts
Stack the deck in your favor when hiring people into your data team.
Data science is all the rage, but some companies focus on hiring just data scientists. Be careful.
Let's walk through the decision of whether your company would benefit from building a Hadoop cluster.
use your customer data to really engage your customers