Apr 30, 2021

Customer Intelligence Platform Affogata Raises $5.5M Funding

AI
Data
Insights
platforms
Tilly Kenyon
2 min
Affogata raises $5.5m for their AI-powered customer intelligence platform, which organisations can use to track and analyse competitors’ strategies
Affogata raises $5.5m for their AI-powered customer intelligence platform, which organisations can use to track and analyse competitors’ strategies...

Affogata, a customer intelligence platform enabling brands to harness data to improve the consumer experience, has raised $5.5M in seed financing. The round was co-led by Mangrove Capital Partners and PICO Venture Partners. Wix Capital and Fiverr Founder and CEO, Micha Kaufman, also participated in the funding round.

This new funding will finance the company’s future expansion plans, including R&D, customer solutions, and more. Affogata's artificial intelligence-driven platform obtains information from a wealth of open data sources about the brand and its competitors to provide real-time actionable insights. The platform's vast array of data sources includes niche sites that have critical importance for specific segments such as gaming forums, fintech communities, and more.

"Our ability to provide customer opinions and actionable insights for multiple stakeholders in real-time is a major differentiator for any brand, especially with the continued rise of online communities like Reddit," said Affogata Co-founder and CEO Sharel Omer. "This funding round is a big vote of confidence in our platform and will provide us with the ability to empower countless brands with a robust set of tools to improve their products, strengthen their online presence, and really stand apart when it comes to offering a superior customer experience while maintaining a strong reputation."

Affogata not only gathers data

Affogata not only gathers data but also makes insights and actions accessible to multiple stakeholders in a company, allowing them to collaborate with the same speed as a startup. Marketers, product managers, data analysts, customer success teams, PR teams, and other relevant stakeholders gain access to relevant insights and collaborate on comprehensive solutions.

The platform replaces multiple tools that are typically only focused on listening or actions, enabling a high level of automation and proactive responses at the ideal moment, triggered by the data. These responses range from ensuring online communities are free of spammers and bots to detecting anomalies and responding in real-time. It saves moderators hours of manual work and customer support issues are resolved quickly.

"Being able to make decisions to improve product development and the customer experience based on expansive, real-time data is now mission-critical and Affogata's product has become a customer intelligence solution for leading global brands," said Claudia De Antoni, Operating Partner at PICO Venture Partners. 

Share article

Jun 17, 2021

Facebook Develops AI to Crackdown on Deepfakes

Facebook
MSU
AI
Deepfakes
3 min
Social media giant, Facebook, has developed artificial intelligence that can supposedly identify and reverse-engineer deepfake images

In light of the large tidal wave of increasingly believable deepfake images and videos that have been hitting the feeds of every major social media and news outlet in recent years, global organisations have started to consider the risk factor behind them. While the majority of deepfakes are created purely for amusement, their increasing sophistication is leading to a very simple question: What happens when a deepfake is produced not for amusement, but for malicious intent on a grander scale? 

 

Yesterday, Facebook revealed that it was also concerned by that very question and that it had decided to take a stand against deepfakes. In partnership with Michigan State University, the social media giant presented “a research method of detecting and attributing deepfakes that relies on reverse engineering from a single AI-generated image to the generative model used to produce it.” 

 

The promise is that Facebook’s method will facilitate deepfake detection and tracing in real-world settings, where the deepfake image itself is often the only information detectors have to work with. 

Why Reverse Engineering? 

Right now, researchers identify deepfakes through two primary methods: detection, which distinguishes between real and deepfake images, and image attribution, which identifies whether the image was generated using one of the AI’s training models. But generative photo techniques have advanced in scale and sophistication over the past few years, and the old strategies are no longer sufficient. 

 

First, there are only so many images presented in AI training. If the deepfake was generated by an unknown, alternative model, even artificial intelligence won’t be able to spot it—at least, until now. Reverse engineering, common practice in machine learning (ML), can uncover unique patterns left by the generating model, regardless of whether it was included in the AI’s training set. This helps discover coordinated deepfake attacks or other instances in which multiple deepfakes come from the same source. 

 

How It Works 

Before we could use deep learning to generate images, criminals and other ill-intentioned actors had a limited amount of options. Cameras only had so many tools at their disposal, and most researchers could easily identify certain makes and models. But deep learning has ushered in an age of endless options, and as a result, it’s grown increasingly difficult to identify deepfakes.

 

To counteract this, Facebook ran deepfakes through a fingerprint estimation network (FEN) to estimate some of their details. Fingerprints are essentially patterns left on an image due to manufacturing imperfections, and they help identify where the image came from. By evaluating the fingerprint magnitude, repetition frequency, and symmetrical frequency, Facebook then applied those constraints to predict the model’s hyperparameters. 

 

What are hyperparameters? If you imagine a generative model as a car, hyperparameters are similar to the engine components: certain properties that distinguish your fancy automobile from others on the market. ‘Our reverse engineering technique is somewhat like recognising [the engine] components of a car based on how it sounds’, Facebook explained, ‘even if this is a new car we’ve never heard of before’. 

 

What Did They Find? 

‘On standard benchmarks, we get state-of-the-art results’, said Facebook research lead Tal Hassner. Facebook added that the fingerprint estimation network (FEN) method can be used for not only model parsing, but detection and image attribution. While this research is the first of its kind, making it difficult to assess the results, the future looks promising. 


Facebook’s AI will introduce model parsing for real-world applications, increasing our understanding of deepfake detection. As cybersecurity attacks proliferate, and generative AI falls into the hands of those who would do us harm, this method could help the ‘good guys’ stay one step ahead. As Hassner explained: ‘This is a cat-and-mouse game, and it continues to be a cat-and-mouse game’.

Share article