Technology

In advance of JumiaNovember financial reporting, itworth revisiting the companysecond quarter results, the downside of which included some negative news beyond losses.

The Africa focused e-commerce company — with online verticals in 14 countries — did post second-quarter revenue growth of 58% (≈$43 million) and increased its customer base to 4.8 million from 3.2 million over the same period a year ago.

But Jumia also posted greater losses for the period, €67.8 million, compared to €42.3 million in 2018.

What appears to have struck the market more than revenues or losses was Jumia offering greater detail on the fraud perpetrated by some employees and agents of its JForce sales program.

This was another knock for the firm on its up and down ride since becoming the first tech company operating in Africa to list on the NYSE in April. The online retailer gained investor confidence out of the gate, more than doubling its $14.95 opening share price after the IPO.

That lasted until May, when Jumiastock came under attack from short-seller Andrew Left, whose firm Citron Research, issued a report accusing the company of fraud. That prompted several securities related lawsuits against Jumia.

At quick glance, Citronprimary claim — that JumiaSEC filing contained discrepancies in sales figures — shares some resemblance to Jumiaown disclosures.

The companyshare-price has suffered due to both — falling to less than 50% of its opening in April.

This has all funneled into an ongoing debate across Africatech ecosystem on Jumialegitimacy as an African startup, given its (primarily) European senior management. Some of the most critical voices have gone so far as to support Leftclaims on Jumiafraud — and accept JumiaAugust admission as validation.

Sound messy and confusing? We&ll, yes, it is. But so go some IPOs.

Jumiainfo vs. Citronclaims

Evaluating JumiaJ-Force scandal vs. Citronshort-sell claims is really Chartered Financial Analyst stuff. Citibank Research issued a brief rebutting Leftclaims in May and then another in August — though the firm has not made either public.

Judging by Jumiashare-price fluctuation and chatter that continues in Africatech ecosystem, therestill confusion around both matters.

A simple exercise is to lay out the core of what Jumia has released vs. the crux of Citron Research claims.

On the J-Force/improper sales matter, here are excerpts of Jumiastatement. Note that GMV is Gross Merchandise Value— the total amount of goods sold over the period:

Asdisclosed in our prospectus dated April 11, 2019, we received information alleging that some of our independent sales consultants, members of our JForce program in Nigeria, may have engaged in improper sales practices. In response, we launched a review of sales practices covering all our countries of operation and data from January 1, 2017 to June 30, 2019.

Jumia did disclose this in its IPO prospectus on page 34.

In the course of this review, we identified several JForce agents and sellers who collaborated with employees in order to benefit from differences between commissions charged to sellers and higher commissions paid to JForce agents. The transactions in question generated approximately 1% of our GMV in each of 2018 and the first quarter of 2019 and had virtually no impact on our 2018 or 2019 financial statements. We have terminated the employees and JForce agents involved, removed the sellers implicated and implemented measures designed to prevent similar instances in the future. The review of this matter is closed.

And finally, Jumia noted this:

More recently, we have also identified instances where improper orders were placed, including through the JForce program, and subsequently cancelled. Based on our findings to date, we believe that the transactions in question generated approximately 2% of our GMV in 2018, concentrated in the fourth quarter of 2018, approximately 4% in the first quarter of 2019 and approximately 0.1% in the second quarter of 2019. These 0.1% have already been adjusted for in the reported GMV figure for the second quarter of 2019. These transactions had no impact on our financial statements. We have suspended the employees involved pending the outcome of our review and are implementing measures designed to prevent similar instances in the future. We continue our review of this matter.

Thatthe gist of Jumiadisclosure: a small number of employees cooked some sales numbers and commissions, it was negligible to our financials, we flagged the investigation in our IPO prospectus, we took action, we ended it.

The Citron Research report Andrew Left issued to support his short-sell position made several critical claims regarding Jumia, but labeled &the smoking gun& as alleged material inconsistencies between an October 2018, Jumia investor presentation and JumiaApril SEC Form F-1.

For the year 2017, therea difference of 600,000 active customers and 10,000 merchants in Jumiareporting between the fall 2018 investor presentation and the recent 2019 F-1, according to Citron Research. Citron also goes on to press concerns with GMV:

In order to raise more money from investors, Jumia inflated its active consumers and active merchants figures by 20-30% (FRAUD).

The most disturbing disclosure that Jumia removed from its F-1 filing was that 41% of orders were returned, not delivered, or cancelled.

This was previously disclosed in the CompanyOctober 2018 confidential investor presentation. This number is so alarming that is screams fraudulent activities. Instead, Jumia disclosed that &orders accounting for 14.4% of our GMV were either failed deliveries or returned by our consumers& in 2018.

TechCrunch connected with JumiaCEO Sacha Poignonnec and Citron ResearchAndrew Left since the August earnings reporting and disclosures.

On whether Jumiarevelation of improper sales practices validated the fraud claims in CitronBrief, &Itnot the same,& Poignnonec,& told me on a call last month.

&For every one of those allegations,& he said referring to Leftresearch, &there is a clear and simple answer for each of them and we have provided those,& said Poignnonec.

Where is Andrew Left on the matter?&I&m no longer short the stock& he told TechCrunch in a mail this week.

&But that does not mean the stock is a buy whatsoever,& he added — sticking to the fundamentals of his May brief.

What to make of it all?

It appears that what Jumia disclosed in its April prospectus (and added more detail to in August) does not provide one-to-one validation of the claims in Citron ResearchMay report.

But then again, the entire matter — the data, the similar terminology, the multiple docs and disclosures — is still all a bit confusing.

That was evident in an exchange between Sacha Poignonnec and CNBC contributor John Fortt after Jumia2nd quarter earnings call (see 1:19). Fort pressed Poignonnec on Leftclaims vs. Jumiaadmissions and still came away a bit puzzled.

The market, too, appears to be impacted by the fuzziness around Jumiadisclosure of improper sales practices and Andrew Leftclaims.

Jumiashare price plummeted 43% the week Left released his short-sell claims, from $49 to $26.

The companystock price has continued to decline since JumiaAugust earnings call (and sales-fraud disclosure) to $6.52 at close Monday.

That50% below the companyopening in April and 80% below its high before CitronResearch brief and Andrew Leftshort-sell position.

Jumiacore investors appeared to show continued confidence in the company this month, when there wasn&t a big selloff after the IPO lockup period expired.

Even so, Jumia3rd quarter earningcall on November 12 could be a bit make or break for the company with investors given all the volatility the e-commerce venture has faced since listing and its rapid loss in value.

As a public company now, the most direct way for Jumia to revive its share-price (and investor confidence) would be demonstrating it has reduced losses while maintaining or boosting revenues.

Of course, thatthe prescription for just about any recently IPO&d tech venture.

What Jumia may want to evaluate pre-earnings call is the extent to which its own sales-fraud disclosure and Andrew Leftallegations are still being mashed together and impacting brand-equity in Africa and investor confidence abroad.

From there it could be wise to address both head on and explain — in a way that is as easy as possible for people to understand — how the two are not the same and don&t have a bearing on Jumiabrand or business model.

A report by campaign group Avaaz examining how Facebookplatform is being used to spread hate speech in the Assam region of North East India suggests the company is once again failing to prevent its platform from being turned into a weapon to fuel ethnic violence.

Assam has a long-standing Muslim minority population but ethnic minorities in the state look increasingly vulnerable after IndiaHindu nationalist government pushed forward with a National Register of Citizens (NRC), which has resulted in the exclusion from that list of nearly 1.9 million people — mostly Muslims — putting them at risk of statelessness.

In July the United Nations expressed grave concern over the NRC process, saying therea risk of arbitrary expulsion and detention, with those those excluded being referred to Foreigners& Tribunals where they have to prove they are not &irregular&.

At the same time, the UN warned of the rise of hate speech in Assam being spread via social media — saying this is contributing to increasing instability and uncertainty for millions in the region. &This process may exacerbate the xenophobic climate while fuelling religious intolerance and discrimination in the country,& it wrote.

Therean awful sense of deja-vu about these warnings. In March 2018 the UN criticized Facebook for failing to prevent its platform being used to fuel ethnic violence against the Rohingya people in the neighboring country of Myanmar — saying the service had played a &determining role& in that crisis.

Facebook response to devastating criticism from the UN looks like wafer-thin crisis PR to paper over the ethical cracks in its ad business, given the same sorts of alarm bells are being sounded again, just over a year later. (If we measure the company by the lofty goals it attached to a director of human rights policy job last year — when Facebook wrote that the responsibilities included &conflict prevention& and &peace-building& — itsurely been an abject failure.)

Avaazreport on hate speech in Assam takes direct aim at Facebookplatform, saying itbeing used as a conduit for whipping up anti-Muslim hatred.

In the report, entitled Megaphone for Hate: Disinformation and Hate Speech on Facebook During AssamCitizenship Count, the group says it analysed 800 Facebook posts and comments relating to Assam and the NRC, using keywords from the immigration discourse in Assamese, assessing them against the three tiers of prohibited hate speech set out in FacebookCommunity Standards.

Avaaz found that at least 26.5% of the posts and comments constituted hate speech. These posts had been shared on Facebook more than 99,650 times — adding up to at least 5.4 million views for violent hate speech targeting religious and ethnic minorities, according to its analysis.

Bengali Muslims are a particular target on Facebook in Assam, per the report, which found comments referring to them as &criminals,& &rapists,& &terrorists,& &pigs,& and &dogs&, among other dehumanizing terms.

In further disturbing comments there were calls for people to &poison& daughters, and legalise female foeticide, as well as several posts urging &Indian& women to be protected from &rape-obsessed foreigners&.

Avaaz suggests its findings are just a drop in the ocean of hate speech that it says is drowning Assam via Facebook and other social media. But it accuses Facebook directly of failing to provide adequate human resource to police hate speech spread on its dominant platform.

Commenting in a statement, Alaphia Zoyab, senior campaigner, said: &Facebook is being used as a megaphone for hate, pointed directly at vulnerable minorities in Assam, many of whom could be made stateless within months. Despite the clear and present danger faced by these people, Facebook is refusing to dedicate the resources required to keep them safe. Through its inaction, Facebook is complicit in the persecution of some of the worldmost vulnerable people.&

Its key complaint is that Facebook continues to rely on AI to detect hate speech which has not been reported to it by human users — using its limited pool of (human) content moderator staff to review pre-flagged content, rather than proactively detect it.

Facebook founder Mark Zuckerberg has previously said AI has a very long way to go to reliably detect hate speech. Indeed, hesuggested it may never be able to do that.

In April 2018 he told US lawmakers it might take five to ten years to develop &AI tools that can get into some of the linguistic nuances of different types of content to be more accurate, to be flagging things to our systems&, while admitting: &Today we&re just not there on that.&

That sums to an admission that in regions such as Assam — where inter-ethnic tensions are being whipped up in a politically charged atmosphere thatalso encouraging violence — Facebook is essentially asleep on the job. The job of enforcing its own ‘Community Standards& and preventing its platform being weaponized to amplify hate and harass the vulnerable, to be clear.

Avaaz says it flagged 213 of &the clearest examples& of hate speech which it found directly to Facebook — including posts from an elected official and pages of a member of an Assamese rebel group banned by the Indian Government. The company removed 96 of these posts following its report.

It argues there are similarities in the type of hate speech being directed at ethnic minorities in Assam via Facebook and that which targeted at Rohingya people in Myanmar, also on Facebook, while noting that the context is different. But it did also find hateful content on Facebook targeting Rohingya people in India.

It is calling on Facebook to do more to protect vulnerable minorities in Assam, arguing it should not rely solely on automated tools for detecting hate speech — and should instead apply a &human-led ‘zero tolerance& policy& against hate speech, starting by beefing up moderators& expertise in local languages.

It also recommends Facebook launch an early warning system within its Strategic Response team, again based on human content moderation — and do so for all regions where the UN has warned of the rise of hate speech on social media.

&This system should act preventatively to avert human rights crises, not just reactively to respond to offline harm that has already occurred,& it writes.

Other recommendations include that Facebook should correct the record on false news and disinformation by notifying and providing corrections from fact-checkers to each and every user who has seen content deemed to have been false or purposefully misleading, including if the disinformation came from a politician; that it should be transparent about all page and post takedowns by publishing its rational on the Facebook Newsroom so the issue of hate speech is given proportionate prominence and publicity to the size of the problem on Facebook; and it should agree to an independent audit of hate speech and human rights on its platform in India.

&Facebook has signed up to comply with the UN Guiding Principles on Business and Human Rights,& Avaaz notes. &Which require it to conduct human rights due diligence such as identifying its impact on vulnerable groups like women, children, linguistic, ethnic and religious minorities and others, particularly when deploying AI tools to identify hate speech, and take steps to subsequently avoid or mitigate such harm.&

We reached out to Facebook with a series of questions about Avaazreport and also how it has progressed its approach to policing inter-ethnic hate speech since the Myanmar crisis — including asking for details of the number of people it employs to monitor content in the region.

Facebook did not provide responses to our specific questions. It just said it does have content reviewers who are Assamese and who review content in the language, as well as reviewers who have knowledge of the majority of official languages in India, including Assamese, Hindi, Tamil, Telugu, Kannada, Punjabi, Urdu, Bengali and Marathi.

In 2017 India overtook the US as the country with the largest &potential audience& for Facebook ads, with 241M active users, per figures it reports the advertisers.

Facebook also sent us this statement, attributed to a spokesperson:

We want Facebook to be a safe place for all people to connect and express themselves, and we seek to protect the rights of minorities and marginalized communities around the world, including in India. We have clear rules against hate speech, which we define as attacks against people on the basis of things like caste, nationality, ethnicity and religion, and which reflect input we received from experts in India. We take this extremely seriously and remove content that violates these policies as soon as we become aware of it. To do this we have invested in dedicated content reviewers, who have local language expertise and an understanding of the Indialongstanding historical and social tensions. We&ve also made significant progress in proactively detecting hate speech on our services, which helps us get to potentially harmful content faster.

But these tools aren&t perfect yet, and reports from our community are still extremely important. Thatwhy we&re so grateful to Avaaz for sharing their findings with us. We have carefully reviewed the content they&ve flagged, and removed everything that violated our policies. We will continue to work to prevent the spread of hate speech on our services, both in India and around the world.

Facebook did not tell us exactly how many people it employs to police content for an Indian state with a population of more than 30 million people.

Globally the company maintains it has around 35,000 people working on trust and safety, less than half of whom (~15,000) are dedicated content reviewers. But with such a tiny content reviewer workforce for a global platform with 2.2BN+ users posting night and day all around the world thereno plausible no way for it to stay on top of its hate speech problem.

Certainly not in every market it operates in. Which is why Facebook leans so heavily on AI — shrinking the cost to its business but piling content-related risk onto everyone else.

Facebook claims its automated tools for detecting hate speech have got better, saying that in Q1 this year it increased the proactive detection rate for hate speech to 65.4% — up from 58.8% in Q4 2017 and 38% in Q2 2017.

However it also says it only removed 4 million pieces of hate speech globally in Q1. Which sounds incredibly tiny vs the size of Facebookplatform and the volume of content that will be generated daily by its millions and millions of active users.

Without tools for independent researchers to query the substance and spread of content on Facebookplatform itsimply not possible to know how many pieces of hate speech are going undetected. But — to be clear — this unregulated company still gets to mark its own homework.

In just one example of how Facebook is able to shrink perception of the volume of problematic content itfencing, of the 213 pieces of content related to Assam and the NCR that Avaaz judged to be hate speech and reported to Facebook it removed less than half (96).

Yet Facebook also told us it takes down all content that violates its community standards — suggesting it is applying a far more dilute definition of hate speech than Avaaz. Unsurprising for a US company whose nascent crisis PR content review board‘s charter includes the phrase &free expression is paramount&. But for a company that also claims to want to prevent conflict and peace-build itrather conflicted, to say the least.

As things stand, Facebookself-reported hate speech performance metrics are meaningless. Itimpossible for anyone outside the company to quantify or benchmark platform data. Because no one except Facebook has the full picture — and itnot opening its platform for ethnical audit. Even as the impacts of harmful, hateful stuff spread on Facebook continue to bleed out and damage lives around the world.

While large entertainment companies scramble to catch up to streaming content platforms, more fundamental upheaval is headed their way as a result of technological advances in artificial intelligence and 5G.

Former ProSiebenSat.1 executive Kevin Baxpehler (based in Tel Aviv) and former Google Ventures partner Eze Vidra (based in London) launched Remagine Ventures earlier this year with a $35 million fund that bridges the gap between technologists at the forefront of change and the largest owners of content.

Backed by a roster of multi-billion-dollar media companies in Europe, Asia and the U.S. as its limited partners, their firm operates independently (and focuses on financial return) but aims to provide strategic value to portfolio companies and insight into the future for its LPs. Vidra referred to it as &a multi-corporate Google Ventures type of model.&

The firmfocus on entertainment technologies has a B2B bent, with a geographic focus on Israel as its primary hub and with most of its initial portfolio selling to enterprise media companies. That makes Remagineability to guide entrepreneurs through the halls of traditional media giants highly relevant; it also means it can gauge whether traditional media companies are likely to buy a startupproduct before they invest.

I spoke with Baxpehler and Vidra to learn more about their playbook and why they believe a wave of entertainment tech companies is about to come out of Israel. Herethe transcript of our conversation (edited for length and clarity):

Eric Peckham: Are there specific investment theses within entertainment that you are hunting for startups in?

Kevin Baxpehler: Our investment thesis is based on two main drivers: new advancements in so-called AI technologies — specifically deep-learning, computer-vision and NLP — coupled with new consumer trends such as esports, visual search, and engaging with computer-generated imagery (CGI) like Lil Miquela.

We believe that recent technological developments such as GANs (generative adversarial networks), coupled with new powerful computing power like new microprocessing chips and 5G, will change how brands, consumers, and stars/influencers will all interact. It creates tremendous opportunities to invest.

Eze Vidra: Remagine Ventures invests independently in seed and pre-seed startups at the intersection of entertainment, tech, data and commerce. Seed investing is particularly hard for corporates to do directly (because of a combination of reasons including speed, signaling risk and the challenges of deal flow for corporates) so we specialise at that stage by sourcing real time feedback from the market.

We are seeing industries and disciplines converge and find the intersections to be the most ripe areas of opportunity. For example, content + commerce, AI + entertainment, gaming + live stream tech giving us esports as a cultural phenomenon changing consumer behaviour.

Give me some examples of what startups at these intersection points will look like.

Vidra: The two core tenants of our thesis are 1) changing consumer behavior — for example, how esports is moving young viewers to engage with gaming — and 2) new technologies that make new forms of entertainment possible, primarily driven by AI.

Our portfolio company Syte is an image-recognition and computer-vision company that recognizes the products inside images and videos with a very high degree of accuracy. They are working with top retailers globally and Samsung selected them to power the Bixby assistant and is rolling them out globally. Itbeen tried before, but the difference with Syteproduct is the level of accuracy.

We invested in HourOne, which is a synthetic video company using generative adversarial networks to generate video without the camera. It has multiple use cases, from reducing the cost of video production to programmatic video, to text-to-speech to gaming.

Another example is Vault, which uses deep learning to predict the success of scripted projects, whether itmovies or TV shows down to the box office opening Rotten Tomatoes scores, the probability of there being a season two, the demographics that are most impacted, etc. So bringing a more data-driven approach to marketing films and shows.

Being vertically-focused means that we can attract relevant dealflow from both entrepreneurs and co-investors. As we evaluate startups, we look for interesting teams that are leveraging new technology (or taking an interesting consumer angle) that can scale and we focus on helping them open doors internationally.

To what extent is your interest focused on startups selling their technology to enterprise media companies versus startups building tools for the broader landscape of small content creators?