loader from loading.io

Episode 86 - 3D Printing, AI Analysis of Reddit, and Bias in Facial Recognition

The Killing IT Podcast

Release Date: 11/17/2020

Episode 112 - AI in the Military, Colonial Pipeline, and Discriminatory NDAs show art Episode 112 - AI in the Military, Colonial Pipeline, and Discriminatory NDAs

The Killing IT Podcast

Topic 1: Military practices controlling hundreds of robots and drones at once - with AI Like it or not, AI is coming to military operations. It might be cheaper, but is it safer? But is the reliance on AI a problem? Links:     Topic 2: What can we learn from the Colonial Pipeline hack? How should companies respond? And are there real consequences for companies that get hacked? We discuss. Topic 3: Proposed law says discriminatory behavior should not be subject to non-disclosure agreements. Is the NDA … too much? Ifeoma Ozoma is a technology policy expert with experience leading...

info_outline
Episode 111 - Hacking Waze, Cookies by Drone, and Discussing Diversity show art Episode 111 - Hacking Waze, Cookies by Drone, and Discussing Diversity

The Killing IT Podcast

Topic 1: Burger King diverts McDonalds customers - via Waze This one is kinda cool because it's so sneaky. With just a taste of big data. When people are close to a McDonalds billboard and headed toward a restaurant, Waze lures them off to a Burger King instead. Just sneaky, or actually immoral? Link: Topic 2: Girl Scout Cookies by Drone Not a burrito or pizza. But it's pretty cool to get your cookies by drone. Makes us ponder: Are there things you would actually pay a premium to get via drone? We're thinking yes for cookies, and probably no for a ream of paper. Link: Topic 3: Basecamp...

info_outline
Episode 110 - Hacked Ice Cream Machine and Right to Repair show art Episode 110 - Hacked Ice Cream Machine and Right to Repair

The Killing IT Podcast

Topic 1: Hacked Ice Cream Makers at McDonalds A very strange right to repair! If you buy an expensive machine, do you have the right to buy a third-party add-on that makes it work better? Topic 2: EU looking to regulate “High-Risk” AI High risk determined in two steps by considering whether it may cause certain harms and, if so, the severity of the possible harm and the probability of occurrence. Possible damages include - death of a person - damage of property - systemic adverse impacts for society at large - adverse impact on financial, educational or professional opportunities of...

info_outline
Episode 109 - Offensive Tech Jargon, Facial Recognition Update, and Geo-Location Ethics show art Episode 109 - Offensive Tech Jargon, Facial Recognition Update, and Geo-Location Ethics

The Killing IT Podcast

Topic 1: Racist (and other offensive) Tech Jargon Words matter. The most obvious example is "master/slave" but there's plenty of other insensitive terminology around. Remember, these terms might make sense to you, but you clients may hear it and wonder exactly what you're talking about. Link: Topic 2: Facial Recognition Update. Facial Recognition has evolved considerably in the pandemic. It has been banned in a few places. It has been regulated a bit at the local (not federal) level. Amateurs are now getting in the game (e.g., Used to identify police during protests). And the police have...

info_outline
Episode 108 - AI on Fiverr, Millions of IOT Devices Compromised, and More show art Episode 108 - AI on Fiverr, Millions of IOT Devices Compromised, and More

The Killing IT Podcast

Topic 1: Will Bots Take Jobs? AI Robots are coming for your jobs! Or - in a much more realistic interpretation, they are coming to reduce the amount of mundane process work that humans do so they can focus on interpersonal stuff that robots can’t do. We take a mostly positive spin on this one. Related Link: Topic 2: 100 Million IOT Devices Compromised Scary fact: As the number of deployed devices continues to grow exponentially, the size of these incidents will also continue to grow. This will also have an effect on our liability insurance and state/provincial legislation. Related Link: ...

info_outline
Episode 107 - Best Practices, Ingram/Harmony, and Water Cooled Servers show art Episode 107 - Best Practices, Ingram/Harmony, and Water Cooled Servers

The Killing IT Podcast

Topic 1: Should you rely on best practices? We return to the question of what ARE best practices and how you define them. But more than that, should you be following other peoples' so-called best practices? Stop leaning on best practices, and experiment! Related Link: Topic 2: Distributor Ingram Micro Buys Harmony PSA. First, do you care? Second, is this a good move? Would you be tempted to a PSA owned by one of your distributors? Related Links: Topic 3: Submerged Servers. Like, IN the water. Last year Microsoft completed a test of sealed server data centers at the bottom of the ocean....

info_outline
Episode 106 - Multicloud, Deep Fakes, AI, and More show art Episode 106 - Multicloud, Deep Fakes, AI, and More

The Killing IT Podcast

Topic 1: Stop Talking About Multicloud and Hybrid Cloud It's time to take this seriously and talk about Integration. As always, we take a stab at the terminology used by our industry and ponder about what's really going on. We all love jargon, but it's important to dig a little a figure out what's really being discussed. Related Link: Topic 2: Detecting Deef Fake Videos with AI Scientists developed a clever way to detect deep fakes by analyzing light reflections in the eyes Related Link: Book to note: Thinking, Fast and Slow by Daniel Kahneman. Topic 3: Next Up - Open AI that doesn’t...

info_outline
Episode 105 - Cops and Robots, Distributors and Bitcoin show art Episode 105 - Cops and Robots, Distributors and Bitcoin

The Killing IT Podcast

Topic 1: Cops and Robots Move to stop NYC police from using armed robots. Note: The police do not own any armed robots, but they do own some robots. So this is totally a preventive measure. We previously discussed the use of AI and facial recognition by police departments. The question here is, when do we begin the conversation (before or after deployment)? Related Link:   Topic 2: Is Bitcoin mining compatible with Singapore’s green initiatives? Bitcoin mining uses about the same amount electricity as Norway! Now Singapore is questioning whether this non-green factor is consistent with...

info_outline
Episode 104 - AI Mayflower, NFT Artwork, and RMM News show art Episode 104 - AI Mayflower, NFT Artwork, and RMM News

The Killing IT Podcast

Topic 1: AI Mayflower to Cross the Atlantic This is a recreation of the Mayflower voyage with a 100% autonomous ship. Great for research and proof of concept. The Smithsonian doesn’t talk about autonomous shipping vessels, but that might be an option going forward. There are currently very small crews on shipping vessels, but there might be some advantages with this technology. Related Link: We update our burrito bet. Topic 2: NFT - Non-fungible Token Artwork By now you've heard of NFTs, particularly artwork stored with Etherium blockchain guarantee of authenticity. Sold on Rarible.com or...

info_outline
Episode 103 - Big Tech Lobbying, Massive Security Camera Breach show art Episode 103 - Big Tech Lobbying, Massive Security Camera Breach

The Killing IT Podcast

Topic 1: What works, works: Big Tech Targets DC via Old-School Media Facebook, Amazon, and Google ads are blanketing inside-the-Beltway newsletters in a bid to rehab their tarnished reputations. The big tech companies are getting ready to do battle. So they need to soften up members of Congress. Here’s the interesting part: They didn’t use social media to get the job done. They used good old advertising . . . in newsletters. Of course we use this to look at the whole "Big Data vs. Congress" debate going on right now. Link:   Topic 2: Security Breach at Security Camera Company IoT...

info_outline
 
More Episodes

Topic 1: 3D printing that makes sense.

We have covered 3D printing several times. Now we get to report on some seriously cool applications. Medical applications are some of the coolest applications for 3D printing.

3d Printing is moving out of the "emerging" phase and finding some real uses in the real world.

https://www.axios.com/newsletters/axios-future-a832ca67-a3a6-46a4-b805-9e9471b3e8a8.html (See story #3)


Topic 2: AI Analysis of Reddit users.

Here's an interesting use of AI. The folks over at MIT (Massachusetts Institute of Technology) analyzed the words and moods of writers on Reddit over time. They found some interesting shifts in personal emotions over time.

We look into the tech as well as the environment (Reddit). There are always questions about the research techniques, but it helps a lot that this is from MIT and not some random blogger.

https://thenextweb.com/neural/2020/11/06/mit-ai-analysis-of-reddit-shows-anxiety-levels-are-sky-high-since-pandemic/


Topic 3: Taking the middle ground on facial recognition and bias.

We've covered the racial biases that are often built into facial recognition. Now, the Detroit city counsel is taking the middle ground: Don't throw it away; but don't use it as it ships (filled with problems).

This is another area where the prices are coming down and the use cases are increasing.

https://www.nytimes.com/2020/11/11/technology/facial-recognition-software-police.html


Sponsor Note: Gozynta Tixt

Headaches with customers texting your technicians? Gozynta Tixt brings your texting conversation into service tickets. Set your MSP apart by allowing your customers to communicate the way they want, without losing control of the conversation. Since text messages are 10 times more likely to be opened than email, you can get your tickets to "Resolved" faster! An essential tool for supporting a growing remote workforce. Go to www.tixt.us to sign up for a free 30 day trial.

:-)