Medium has announced that starting May 1, 2024, AI-generated content will no longer be eligible for its paid Partner program. The platform will remove such content from behind paywalls and authors may risk expulsion from the monetization scheme.

Medium bans AI usage on its Partner Program

“We are writing to notify you of policy updates that may impact your participation in the Medium Partner Program,” Medium stated in an email shared by Jonathan Gillham:,

Google and Medium have both just released updated policies identifying and restricting the use of AI text on their platforms…

April 9th @Google Updated its Merchant Center requiring merchants to identify AI-generated text:

"will require merchants to disclose text content that… pic.twitter.com/EYc5kQ1tza

— Jonathan Gillham (@JonGillhams) April 11, 2024

The company emphasizes its commitment to human-driven narratives and has decided against hosting entirely AI-created stories. While acknowledging the benefits of AI in enhancing clarity and aiding non-native language writers, Medium insists that content should not be solely produced by AI.

“Medium-is-for-human-stocytelling, not AI-generated writing. We recently defined and clarified our specific policies around the different uses of Al-generated content and technologies, and what is allowed in the Medium Partner Program. You can read those policies in our Help Center.”

-Medium

Medium has updated and specified its guidelines regarding AI-generated content, which are detailed in the Help Center. The platform’s existing guidelines already prohibited the distribution of fully AI-generated stories beyond the writer’s immediate circle on Medium.

“Medium is not a place for fully AI-generated stories, and 100% AI-generated stories will not be eligible for distribution beyond the writer’s personal network,” reads the guidelines.

The use of AI tools to assist in storytelling is permitted under Medium’s guidelines, but there are stringent transparency requirements. Any usage of AI must be explicitly acknowledged within the first two paragraphs of the story. This requirement also applies to AI-generated images, which must be appropriately labeled and sourced.

Although AI-generated stories can still be published on individual blogs within Medium, they will not be widely disseminated across the platform. With this new policy, Medium aims to further limit the distribution of AI-generated content. Users who prefer not to see AI-generated stories have the option to click the “Show less like this” button, which helps the platform tailor content to user preferences and restrict the proliferation of fully AI-generated material.

Featured image credit: Pedro Araújo/Unsplash

This is a follow up to an article Per wrote last year, Machine Learning in a Week, on how he kickstarted his way into machine learning (ml) by devoting five days to the subject. Follow him on Medium and check out his archive.

First intro: Hacker News and Udacity

My interest in ml stems back to 2014 when I started reading articles about it on Hacker News. I simply found the idea of teaching machines stuff by looking at data appealing. At the time I wasn’t even a professional developer, but a hobby coder who’d done a couple of small projects.

So I began watching the first few chapters of Udacity’s Supervised Learning course, while also reading all articles I came across on the subject.

This gave me a little bit of conceptual understanding, though no practical skills. I also didn’t finish it, as I rarely do with MOOC’s.

Failing Coursera’s ML Course

In January 2015 I joined the Founders and Coders (FAC) bootcamp in London in order to become a developer. A few weeks in, I wanted to learn how to actually code machine learning algorithms, so I started a study group with a few of my peers. Every Tuesday evening, we’d watch lectures from Coursera’s Machine Learning course.

It’s a fantastic course, and I learned a hell of a lot. But it’s tough for a beginner. I had to watch the lectures over and over again before grasping the concepts. The Octave coding task are challenging as well, especially if you don’t know Octave. As a result of the difficulty, one by one fell off the study group as the weeks passed. Eventually, I fell off it myself as well.

I hindsight, I should have started with a course that either used ml libraries for the coding tasks — as opposed to building the algorithms from scratch — or at least used a programming language I knew.

Learning a new language while also trying to code ml algorithms is too hard for a newbie.

If I could go back in time, I’d choose Udacity’s Intro to Machine Learning, as it’s easier and uses Python and Scikit Learn. This way, we would have gotten our hands dirty as soon as possible, gained confidence, and had more fun.

Lesson learned: Start with something easy and practical rather than difficult and theoretical.

Machine Learning in a Week

One of the last things I did at FAC was the ml week stunt. My goal was to be able to apply machine learning to actual problems at the end of the week, which I managed to do.

Throughout the week I did the following:

• got to know Scikit Learn
• tried ml on a real world dataset
• coded a linear regression algorithm from scratch (in Python)
• did a tiny bit of nlp

It’s by far the steepest ml learning curve I’ve ever experienced. Go ahead andread the article if you want a more detailed overview.

Lesson learned: Setting off a week solely to immerse yourself into a new subject is extremely effective.

Failing neural networks

After I finished FAC in London and moved back to Norway, I tried to repeat the success from the ml week, but for neural networks instead.

This failed.

There were simply too many distractions to spend 10 hours of coding and learning every day. I had underestimated how important it was to be surrounded by peers at FAC.

Lesson learned: Find an thriving environment to surround yourself with when doing these kinds of learning stunts.

However, I got started with neural nets at least, and slowly started to grasp the concept. By July I managed to code my first net. It’s probably the crappiest implementation ever created, and I actually find it embarrassing to show off. But it did the trick; I proved to myself that I understood concepts likebackpropagation and gradient descent.

In the second half of the year, my progression slowed down, as I started a new job. The most important takeaway from this period was the leap from non-vectorized to vectorized implementations of neural networks, which involved repeating linear algebra from university.

By the end of the year I wrote an article as a summary of how I learned this.

Testing out Kaggle Contests

During the christmas vacation of 2015, I got a motivational boost again and decided try out Kaggle. So I spent quite some time experimenting with various algorithms for their Homesite Quote Conversion, Otto Group Product Classification and Bike Sharing Demand contests.

The main takeaway from this was the experience of iteratively improving the results by experimenting with the algorithms and the data.

I learned to trust my logic when doing machine learning.

If tweaking a parameter or engineering a new feature seems like a good idea logically, it’s quite likely that it actually will help.

Setting up a learning routine at work

Back at work in January 2016 I wanted to continue in the flow I’d gotten into during Christmas. So I asked my manager if I could spend some time learning stuff during my work hours as well, which he happily approved.

Having gotten a basic understanding of neural networks at this point, I wanted to move on to deep learning.

Udacity’s Deep Learning

My first attempt was Udacity’s Deep Learning course, which ended up as a big disappointment. The contents of the video lectures are good, but they are too short and shallow to me.

And the IPython Notebook assignments ended up being too frustrating, as I spent most of my time debugging code errors, which is the most effective way to kill motivation. So after doing that for a couple of sessions at work, I simply gave up.

To their defense, I’m a total noob when it comes to IPython Notebooks, so it might not be as bad for you as it was for me. So it might be that I simply wasn’t ready for the course.

Stanford — Deep Learning for NLP

Luckily, I then discovered Stanford’s CS224D and decided to give it a shot. It is a fantastic course. And though it’s difficult, I never end up debugging when doing the problem sets.

Secondly, they actually give you the solution code as well, which I often look at when I’m stuck, so that I can work my way backwards to understand the steps needed to reach a solution.

Though I’ve haven’t finished it yet, it has significantly boosted my knowledge in nlp and neural networks so far.

However it’s been tough. Really tough. At one point, I realized I needed help from someone better than me, so I came in touch with a Ph.D student who was willing to help me out for 40 USD per hour, both with the problem sets as well as the overall understanding. This has been critical for me in order to move on, as he has uncovered a lot of black holes in my knowledge.

Lesson learned: It’s possible to get a good machine learning teacher for around 50 USD per hour. If you can afford it, it’s definitely worth it.

In addition to this, Xeneta also hired a data scientist recently. He’s got a masters degree in math, so I often ask him for help when I’m stuck with various linear algebra an calculus tasks, or ml in general. So be sure to check out which resources you have internally in your company as well.

Boosting Sales at Xeneta

After doing all this, I finally felt ready to do a ml project at work. It basically involved training an algorithm to qualify sales leads by reading company descriptions, and has actually proven to be a big time saver for the sales guys using the tool.

Check out out article I wrote about it below or head over to GitHub to dive straight into the code.

Getting to this point has surely been a long journey. But also a fast one; when I started my machine learning in a week project, I certainly didn’t have any hopes of actually using it professionally within a year.

But it’s 100 percent possible. And if I can do it, so can anybody else.

Like this article? Subscribe to our weekly newsletter to never miss out!

Follow @DataconomyMedia

“The biggest issue for governments today is how to be relevant. If all citizens are treated with dignity and invited to collaborate, it can be easier for administrations to have a direct finger on the pulse of the nation rather than lose it in transmission through multiple layers of bureaucracy”.

This week has seen several landmark announcements at the intersection between big data and bureaucracy. Firstly, we discovered that the first Prime Minister to use big data to secure a win, Narendra Modi, continues to innovate in his position of power. His office is using sophisticated big data analytics techniques on the mygov.in portal to ascertain public opinion in key areas of policy, in a move which has been described as a blueprint for future democracy.

Whilst India move towards big data adoption, the EU continue to be wary. A landmark ruling in the European Parliament this week voted to “break up” Google, and separate its search functions from its other revenue streams. Whilst the European Parliament lack the autonomy to implement this ruling, it sends a strong and clear message to Google that the EU remain disdainful of their business practices. The UN also approved “The Right to Privacy in the Digital Age”, keeping data privacy and ethics firmly on the agenda.

Here’s our pick of the rest of the key data science announcements you may have missed this week.

TOP DATACONOMY ARTICLES

The Data Science Skills Network

As a data scientist, I am usually heads down in numbers, patterns, and code, but as crazy as it sounds, one of the hardest parts of my job is actually describing what I do. There are plenty of resources that offer descriptions and guides on the career of a data scientist. I’ve heard them described as those at the intersection of statistics, hacking abilities, and domain expertise. Or, as data analysts who live in San Francisco.

Why Organizations Need a Data Strategy

One of the most important tasks that a Data Architect is often asked to help with is the creation of an Enterprise Data Strategy. But why is Data Strategy so important and what exactly does it consist of, and lastly why is this a task that a Data Architect should be leading or supporting?                                                                                                                       …   

Top 5 Data Science Incubators and Accelerators

Unfortunately, one great idea does not make a successful company. When it comes to founding a data science startup, a visionary idea is essential, but so is hiring the right team, gaining an in-depth knowledge of your market, and putting in place the best business practices to help your idea flourish. And then of course there’s funding.

TOP DATACONOMY NEWS

Medium Rolls Out Slick Visualisation Tool Charted, No Strings Attached                                                                                                

Publishing platform Medium has open-sourced their proprietary data visualisation tool, Charted, on Monday. Now available for public use, data stored on Dropbox or Google Drive can be linked to charted.co and it returns a visualization of that data made shareable through a permalink.                                                                                                    …                                                                                                                                               …

Moogsoft Unveil Deep Monitoring of Docker and OpenStack on Incident.MOOG Platform

Moogsoft are the self-proclaimed “leading provider of operational intelligence for the new era of software”. To ensure that they’re firmly at the cutting edge of new technology, they’ve enhanced their Incident.MOOG software product – a service which provides deep monitoring support for hype-laden Docker container environments, as well as open source cloud computing platform OpenStack.

Qubole Provides Self-managing Data Platform as a Service on Microsoft Azure

Qubole, a self-service Platform for Big Data Analytics, announced a strategic relationship with Microsoft Azure in a step to make Big Data solutions more accessible to more people on more platforms. Big Data is an elastic workload and the ideal cloud use case. The Qubole Data Service is a great way to drive cloud services revenue while delivering improved time to value Microsoft Azure customers.

TOP UPCOMING EVENTS

2 December, 2014- Machine Learning for Sensory Data Analysis, Australia     

Following the inaugural event successfully held in conjunction with AI’2013 in Dunedin, MLSDA’14 joins PRICAI’14 to provide a forum that aims at bringing researchers on sensory technology, networking, and distributed data mining together in order to respond to the unprecedented research challenges and opportunities resulted from the synergy between big data, machine learning, and wireless sensor networking.

2-3 December, 2014- The Chief Data Officer Summit, New York

Data is the most competitive tool that your organization has in order to maintain relevance and growth in today’s complex environment. The ground-breaking Chief Data Officer Summit covers the latest innovations you need to advance your organization’s data strategy and management. In addition to the wealth of insight into data development, data governance and innovation, the summit will dissect the role of the Chief Data Officer like never before.

TOP DATACONOMY JOBS

Pricing Manager / Analyst, Wayfair   

As an Analyst of Pricing you will be responsible for pricing every product that appears on the website. You will manage the daily operational pricing functions while continually seeking to optimize procedures and test strategies to increase gross profit. If you love diving into deep data sets to identify areas for improvement, and be even more enthusiastic about solving those problems then do not hesitate to apply!

Business Intelligence Analyst, CupoNation

We are currently looking for new talent to join our highly professional and dynamic Business Intelligence team. In your role you will be responsible to design end-to-end solutions that meet our company’s Business Intelligence requirements. This covers the definition and implementation of technical requirements for ETL jobs, creation of new data layers and the optimization and enhancement of the current data warehouse infrastructure.

Publishing platform Medium has open-sourced their proprietary data visualisation tool, Charted, on Monday. Now available for public use, data stored on Dropbox or Google Drive can be linked to charted.co and it returns a visualization of that data made shareable through a permalink.

Built a year ago by the Product Science team at Medium as an internal tool to assist in analyzing data and share findings within the company, it can now be availed at charted.co. Users handling “protected or sensitive data” can run a separate instance on their secure network, much like how Medium has been doing.

Essentially, Charted focuses on the following core principles, reports Medium:

• Charted does not store any data. It only fetches and visualizes what the link provides. It also refetches the data every 30 minutes, so the chart is always up-to-date.
• Charted does not transform or manipulate data. It displays only and exactly what it receives. Any necessary calculations or adjustments must already be reflected in the data.
• Charted is not a formatting tool. It is deliberately sparse in features. Charted focuses on getting from the data to the visualization with the fewest decisions possible.

It has minimum optional features for ease of use, adjusts to any screen size and updates are automatic. The open sourced Charted is available on GitHub, while Medium irons out kinks and is open to feedback.

Read more here.

Follow @DataconomyMedia

(Image credit: Medium)