ARCHITECHT Daily: Superhuman AI gets pushback; Oracle gets innovative; and Engine Yard gets acquired

There's not a lot up top today (the combination of a slow news day and a sick child), but I do want t
ARCHITECHT
ARCHITECHT Daily: Superhuman AI gets pushback; Oracle gets innovative; and Engine Yard gets acquired
By ARCHITECHT • Issue #63
There’s not a lot up top today (the combination of a slow news day and a sick child), but I do want to point to these three stories that piqued my interest:
The myth of a superhuman AI (Backchannel): This post by WIRED co-founder Kevin Kelly could’ve used an editor in places, but it does a nice job explaining why a lot of people don’t believe we’re in imminent danger from AI. In a nutshell: There’s no evidence of exponential growth in machine intelligence, and the idea of AI superintelligence makes some big misconceptions about about intelligence itself.
Oracle plans ‘startup organization’ focused on cloud computing, AI, and VR (InfoWorld): Oracle absolutely should be investing in these areas, and it’s probably smart to expect the results of these innovation centers to drive revenue, at least in certain areas (Oracle is interested in cloud computing, big data analytics, mobile computing, internet of things and cybersecurity, as well as the AI stuff). But … Denver and Reston, Virginia, are far from a lot of the talent.
Engine Yard acquired by Crossover (Engine Yard): If you’re not familiar with Engine Yard, it was one of the original PaaS offerings back in the day, focused on Ruby on Rails. It also was the home of Deis, the PaaS-layer-for-Kubernetes that Microsoft acquired earlier this month. The timing of that acquisition makes a lot more sense now, as Engine Yard itself was probably getting ready to join the Ruby-focused folks at Crossover.

Sponsor: Cloudera
Artificial intelligence
A couple things strike me: (1) Nvidia is making quite a production out of its venture program, GPU Ventures, and (2) none of these companies is working on new hardware architectures. 
See point No. 2. above. The amount of work being put into alternative architectures for AI workloads is pretty remarkable. It’s early to predict a winner—GPUs included.
Microsoft made a bit of noise about this last week, but there weren’t a lot of details about how or why it works. This article helps lay out the case for AI and ML in the database.
More from that epic report+survey put out by the Royal Society this week. To my point earlier this week about AI and regulations, I assume companies might like some lane markers, too.
Cloud and infrastructure
Kudos to Twistlock, which seems like it’s onto something with securing container-based infrastructure. For more, check out my interview with c0-founder and CEO Ben Bernstein.
This shouldn’t be too surprising given the push around simplicity and modularity within every large data center operator. Whether Amazon has built this tech is still a mystery. Spooooky …
Here’s a decent rundown of the companies and investment in the container ecosystem. There are some notable omissions, though, including Mesosphere (my previous employer) and Sysdig, off the top of my head. 
Good. Using government web services can be a painstaking experience. Modern development practices can’t hurt.
gcn.com  •  Share
Remember early flash storage company Violin Memory, which publicly exploded a few years ago? It has a new home inside an investment fund managed by billionaire George Soros.
Sponsor: Marshal.io
All things data
I’ll be honest: I don’t know a lot about Infor, except that former Oracle president Charles Phillips is its CEO. And that this acquisition is a move to take on Oracle.
fortune.com  •  Share
The company, founded by former NSA staff, thinks machine learning models need to be auditable and people using them should have info on what’s inside. It’s a smart idea.
Because monitoring modern software environments is hard, especially when they span various clouds, microservices and systems, and when users want granular detail in real time.
This is a good description from real-time analytics company Metamarkets on its Spark-Kafka-Mesos-Druid-Samza environment for big, fast data. The first two pieces are pretty much a must-have for real-time pipelines at this point.
Did you enjoy this issue?
ARCHITECHT
The most interesting news, analysis, blog posts and research in cloud computing, artificial intelligence and software engineering. Delivered daily to your inbox. Curated by Derrick Harris. Check out the Architecht site at https://architecht.io
Carefully curated by ARCHITECHT with Revue. If you were forwarded this newsletter and you like it, you can subscribe here. If you don't want these updates anymore, please unsubscribe here.