Connect with us

Security

Ex-YouTube engineer: Extreme content? No, it’s algorithms that radicalize people

Published

on

Facebook to test anti-misinformation measures in Singapore elections
Advertisers running campaigns on social or political issues in the country now must confirm their identity and location as Facebook looks to stem misinformation.

In the age of fake news and radicalization, the real enemy is not content itself. It’s the algorithm pushing that content up to the top of users’ ‘recommended’ lists. 

That’s according to software engineer Guillaume Chaslot. He should know: he used to work at YouTube and helped build the tech giant’s recommendation algorithm. 

“We need to understand the difference between freedom of speech and freedom of reach. You’re free to say whatever you want to say – but there shouldn’t be freedom to amplify this,” Chaslot told a conference ahead of the Mozilla Festival weekend in London.

He added that extreme content in itself is not problematic. He is actually in favor of having as much content as possible on current platforms. 

Google’s YouTube has recently come in for criticism for its poor management of potentially harmful content. As a result, it has recently made the removal of videos that violate its policy its number one priority.

At the same time, the platform has to perform a delicate balancing act between content moderation and freedom of speech. In a quarterly letter to YouTubers, CEO Susan Wojcicki wrote: “A commitment to openness is not easy. It sometimes means leaving up content that is outside the mainstream, controversial or even offensive.”

The debate about content moderation is not new. But for Chaslot, the issue now is that companies not only publish content, but apply algorithms to it. 

“Algorithms are built to boost watch time, and that typically happens through viewing increasingly radical videos,” he told ZDNet. 

“Someone could be completely radicalized through viewing hours of YouTube videos on end – and from the perspective of the algorithm, that’s actually jackpot.” 

When he worked at YouTube, he said he raised this issue and suggested including more diverse videos in the platform’s recommendation algorithm.

He was met with skepticism from management, so he left the company and started digging to find out where exactly the algorithm would lead him.

This coincided with the 2016 presidential election in the USA, and his research confirmed that YouTube’s algorithm was pushing users to watch more radical videos.

His results were published last year, with the disclaimer that they could only be partial, since the company withholds from the public any data about which content its algorithm promotes.

“We don’t know how much YouTube promotes radical ideas like terrorism,” said Chaslot. “They are doing better but in the course of history, we have no idea, and we will probably never know.”

YouTube has ramped up efforts to change its recommendation algorithm. This year, it launched a trial in the UK to reduce the spread of what it calls “borderline content” after a similar trial in the US halved the views of such content from recommendation, according to the company.

But this is not enough, according to Chaslot. He added that it is now necessary to create efficient legislation to tackle the issue.

“It is similar to when we realized that tobacco was killing people,” he said. “First, we needed the scientific evidence showing that tobacco is harmful – and now, we need the scientific evidence that YouTube is promoting extremism.”

Only once this evidence is produced can there be growing public awareness of the issue, before legislation is introduced, he said. “We made rules to stop people from smoking in public places, not from smoking altogether,” he pointed out. “Something similar should be done with content.”

But with current laws, nothing forces online platforms to share the data that would enable scientific research in the first place. 

As a result, the world is governed by secret algorithms that decide on 70% of what viewers see on YouTube, and 100% of what they read on Facebook, he argued. Euphemistically, Chaslot described this as “a bit crazy”.

ZDNet has contacted YouTube for comment and will update this article if it receives a response.  



Source link

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Security

Key Criteria for Evaluating Security Information and Event Management Solutions (SIEM)

Published

on

Security Information and Event Management (SIEM) solutions consolidate multiple security data streams under a single roof. Initially, SIEM supported early detection of cyberattacks and data breaches by collecting and correlating security event logs. Over time, it evolved into sophisticated systems capable of ingesting huge volumes of data from disparate sources, analyzing data in real time, and gathering additional context from threat intelligence feeds and new sources of security-related data. Next-generation SIEM solutions deliver tight integrations with other security products, advanced analytics, and semi-autonomous incident response.

SIEM solutions can be deployed on-premises, in the cloud, or a mix of the two. Deployment models must be weighed with regard to the environments the SIEM solution will protect. With more and more digital infrastructure and services becoming mission critical to every enterprise, SIEMs must handle higher volumes of data. Vendors and customers are increasingly focused on cloud-based solutions, whether SaaS or cloud-hosted models, for their scalability and flexibility.

The latest developments for SIEM solutions include machine learning capabilities for incident detection, advanced analytics features that include user behavior analytics (UBA), and integrations with other security solutions, such as security orchestration automation and response (SOAR) and endpoint detection and response (EDR) systems. Even though additional capabilities within the SIEM environment are a natural progression, customers are finding it even more difficult to deploy, customize, and operate SIEM solutions.

Other improvements include better user experience and lower time-to-value for new deployments. To achieve this, vendors are working on:

  • Streamlining data onboarding
  • Preloading customizable content—use cases, rulesets, and playbooks
  • Standardizing data formats and labels
  • Mapping incident alerts to common frameworks, such as the MITRE ATT&CK framework

Vendors and service providers are also expanding their offerings beyond managed SIEM solutions to à la carte services, such as content development services and threat hunting-as-a-service.

There is no one-size-fits-all SIEM solution. Each organization will have to evaluate its own requirements and resource constraints to find the right solution. Organizations will weigh factors such as deployment models or integrations with existing applications and security solutions. However, the main decision factor for most customers will revolve around usability, affordability, and return on investment. Fortunately, a wide range of solutions available in the market can almost guarantee a good fit for every customer.

How to Read this Report

This GigaOm report is one of a series of documents that helps IT organizations assess competing solutions in the context of well-defined features and criteria. For a fuller understanding consider reviewing the following reports:

Key Criteria report: A detailed market sector analysis that assesses the impact that key product features and criteria have on top-line solution characteristics—such as scalability, performance, and TCO—that drive purchase decisions.

GigaOm Radar report: A forward-looking analysis that plots the relative value and progression of vendor solutions along multiple axes based on strategy and execution. The Radar report includes a breakdown of each vendor’s offering in the sector.

Solution Profile: An in-depth vendor analysis that builds on the framework developed in the Key Criteria and Radar reports to assess a company’s engagement within a technology sector. This analysis includes forward-looking guidance around both strategy and product.

Continue Reading

Security

Key Criteria for Evaluating Secure Service Access

Published

on

Since the inception of large-scale computing, enterprises, organizations, and service providers have protected their digital assets by securing the perimeter of their on-premises data centers. With the advent of cloud computing, the perimeter has dissolved, but—in most cases—the legacy approach to security hasn not. Many corporations still manage the expanded enterprise and remote workforce as an extension of the old headquarters office/branch model serviced by LANs and WANs.

Bolting new security products onto their aging networks increased costs and complexity exponentially, while at the same time severely limiting their ability to meet regulatory compliance mandates, scale elastically, or secure the threat surface of the new any place/any user/any device perimeter.

The result? Patchwork security ill-suited to the demands of the post-COVID distributed enterprise.

Converging networking and security, secure service access (SSA) represents a significant shift in the way organizations consume network security, enabling them to replace multiple security vendors with a single, integrated platform offering full interoperability and end-to-end redundancy. Encompassing secure access service edge (SASE), zero-trust network access (ZTNA), and extended detection and response (XDR), SSA shifts the focus of security consumption from being either data center or edge-centric to being ubiquitous, with an emphasis on securing services irrespective of user identity or resources accessed.

This GigaOm Key Criteria report outlines critical criteria and evaluation metrics for selecting an SSA solution. The corresponding GigaOm Radar Report provides an overview of notable SSA vendors and their offerings available today. Together, these reports are designed to help educate decision-makers, making them aware of various approaches and vendors that are meeting the challenges of the distributed enterprise in the post-pandemic era.

How to Read this Report

This GigaOm report is one of a series of documents that helps IT organizations assess competing solutions in the context of well-defined features and criteria. For a fuller understanding consider reviewing the following reports:

Key Criteria report: A detailed market sector analysis that assesses the impact that key product features and criteria have on top-line solution characteristics—such as scalability, performance, and TCO—that drive purchase decisions.

GigaOm Radar report: A forward-looking analysis that plots the relative value and progression of vendor solutions along multiple axes based on strategy and execution. The Radar report includes a breakdown of each vendor’s offering in the sector.

Solution Profile: An in-depth vendor analysis that builds on the framework developed in the Key Criteria and Radar reports to assess a company’s engagement within a technology sector. This analysis includes forward-looking guidance around both strategy and product.

Continue Reading

Security

Key Criteria for Evaluating Edge Platforms

Published

on

Edge platforms leverage distributed infrastructure to deliver content, computing, and security closer to end devices, offloading networks and improving performance. We define edge platforms as the solutions capable of providing end users with millisecond access to processing power, media files, storage, secure connectivity, and related “cloud-like” services.

The key benefit of edge platforms is bringing websites, applications, media, security, and a multitude of virtual infrastructures and services closer to end devices compared to public or private cloud locations.

The need for content proximity started to become more evident in the early 2000s as the web evolved from a read-only service to a read-write experience, and users worldwide began both consuming and creating content. Today, this is even more important, as live and on-demand video streaming at very high resolutions cannot be sustained from a single central location. Content delivery networks (CDNs) helped host these types of media at the edge, and the associated network optimization methods allowed them to provide these new demanding services.

As we moved into the early 2010s, we experienced the rapid cloudification of traditional infrastructure. Roughly speaking, cloud computing takes a server from a user’s office, puts it in a faraway data center, and allows it to be used across the internet. Cloud providers manage the underlying hardware and provide it as a service, allowing users to provision their own virtual infrastructure. There are many operational benefits, but at least one unavoidable downside: the increase in latency. This is especially true in this dawning age of distributed enterprises for which there is not just a single office to optimize. Instead, “the office” is now anywhere and everywhere employees happen to be.

Even so, this centralized, cloud-based compute methodology works very well for most enterprise applications, as long as there is no critical sensitivity to delay. But what about use cases that cannot tolerate latency? Think industrial monitoring and control, real-time machine learning, autonomous vehicles, augmented reality, and gaming. If a cloud data center is a few hundred or even thousands of miles away, the physical limitations of sending an optical or electrical pulse through a cable mean there are no options to lower the latency. The answer to this is leveraging a distributed infrastructure model, which has traditionally been used by content delivery networks.

As CDNs have brought the internet’s content closer to everyone, CDN providers have positioned themselves in the unique space of owning much of the infrastructure required to bring computing and security closer to users and end devices. With servers close to the topological edge of the network, CDN providers can offer processing power and other “cloud-like” services to end devices with only a few milliseconds latency.

While CDN operators are in the right place at the right time to develop edge platforms, we’ve observed a total of four types of vendors that have been building out relevant—and potentially competing—edge infrastructure. These include traditional CDNs, hyperscale cloud providers, telecommunications companies, and new dedicated edge platform operators, purpose-built for this emerging requirement.

How to Read this Report

This GigaOm report is one of a series of documents that helps IT organizations assess competing solutions in the context of well-defined features and criteria. For a fuller understanding consider reviewing the following reports:

Key Criteria report: A detailed market sector analysis that assesses the impact that key product features and criteria have on top-line solution characteristics—such as scalability, performance, and TCO—that drive purchase decisions.

GigaOm Radar report: A forward-looking analysis that plots the relative value and progression of vendor solutions along multiple axes based on strategy and execution. The Radar report includes a breakdown of each vendor’s offering in the sector.

Vendor Profile: An in-depth vendor analysis that builds on the framework developed in the Key Criteria and Radar reports to assess a company’s engagement within a technology sector. This analysis includes forward-looking guidance around both strategy and product.

Continue Reading

Trending