Until recently, Elon Musk was just a wildly successful electric car tycoon and space pioneer. Sure, he was erratic and outspoken, but his global influence was contained and seemingly under control.

But add the ownership of just one media platform, in the form of Twitter—now X—and the maverick has become a mogul, and the baton of the world’s biggest media bully has passed to a new player.

What we can gauge from watching Musk’s stewardship of X is that he’s unlike former media moguls, making him potentially even more dangerous. He operates under his own rules, often beyond the reach of regulators. He has demonstrated he has no regard for those who try to rein him in.

Under the old regime, press barons, from William Randolph Hearst to Rupert Murdoch, at least pretended they were committed to truth-telling journalism. Never mind that they were simultaneously deploying intimidation and bullying to achieve their commercial and political ends.

Musk has no need, or desire, for such pretense because he’s not required to cloak anything he says in even a wafer-thin veil of journalism. Instead, his driving rationale is free speech, which is often code for don’t dare get in my way.

This means we are in new territory, but it doesn’t mean what went before it is irrelevant.

A big bucket of the proverbial

If you want a comprehensive, up-to-date primer on the behavior of media moguls over the past century-plus, Eric Beecher has just provided it in his book “The Men Who Killed the News.”

Alongside accounts of people like Hearst in the United States and Lord Northcliffe in the United Kingdom, Beecher quotes the notorious example of what happened to John Major, the UK prime minister between 1990 and 1997, who balked at following Murdoch’s resistance to strengthening ties with the European Union.

In a conversation between Major and Kelvin MacKenzie, editor of Murdoch’s best-selling English tabloid newspaper, The Sun, the prime minister was bluntly told: “Well John, let me put it this way. I’ve got a large bucket of shit lying on my desk and tomorrow morning I’m going to pour it all over your head.”

MacKenzie might have thought he was speaking truth to power, but in reality he was doing Murdoch’s bidding, and actually using his master’s voice, as Beecher confirms by recounting an anecdote from early in Murdoch’s career in Australia.

In the 1960s, when Murdoch owned The Sunday Times in Perth, he met Lang Hancock (father of Gina Rinehart) to discuss potentially buying some mineral prospects together in Western Australia. The state government was opposed to the planned deal.

Beecher cites Hancock’s biographer, Robert Duffield, who claimed Murdoch asked the mining magnate, “If I can get a certain politician to negotiate, will you sell me a piece of the cake?” Hancock said yes. Later that night, Murdoch called again to say the deal had been done. How, asked an incredulous Hancock. Murdoch replied: “Simple […] I told him: look you can have a headline a day or a bucket of shit every day. What’s it to be?”

Between Murdoch in the 1960s and MacKenzie in the 1990s came Mario Puzo’s The Godfather with Don Corleone, aided by Luca Brasi holding a gun to a rival’s head, saying “either his brains or his signature would be on the contract.”

Changing the rules of the game

Media moguls use metaphorical bullets. Those relatively few people who do resist them, like Major, get the proverbial poured over their government. Headlines in The Sun following the Conservatives’ win in the 1992 election included: “Pigmy PM,” “Not up to the job” and “1,001 reasons why you are such a plonker John.”

If media moguls since Hearst and Northcliffe have tap-danced between producing journalism and pursuing their commercial and political aims, they have at least done the former, and some of it has been very good.

The leaders of the social media behemoths, by contrast, don’t claim any fourth estate role. If anything, they seem to hold journalism with tongs as far from their face as possible.

They do possess enormous wealth though. Apple, Microsoft, Google and Meta, formerly known as Facebook, are in the top 10 companies globally by market capitalization. By comparison, News Corporation’s market capitalization now ranks at 1,173 in the world.

Regulating the online environment may be difficult, as Australia discovered this year when it tried, and failed, to stop X hosting footage of the Wakeley Church stabbing attacks. But limiting transnational media platforms can be done, according to Robert Reich, a former Secretary of Labor in Bill Clinton’s government.

Despite some early wins through Australia’s News Media Bargaining Code, big tech companies habitually resist regulation. They have used their substantial influence to stymie it wherever and whenever nation-states have sought to introduce it.

Meta’s founder and chief executive, Mark Zuckerberg, has been known to go rogue, as he demonstrated in February 2021 when he protested against the bargaining code by unilaterally closing Facebook sites that carried news. Generally, though, his strategy has been to deploy standard public relations and lobbying methods.

But his rival Musk uses his social media platform, X, like a wrecking ball.

Musk is just about the first thing the average X user sees in their feed, whether they want to or not. He gives everyone the benefit of his thoughts, not to mention his thought bubbles. He proclaims himself a free-speech absolutist, but most of his pronouncements lean hard to the right, providing little space for alternative views.

Some of his tweets have been inflammatory, such as him linking to an article promoting a conspiracy theory about the savage attack on Paul Pelosi, husband of the former US Speaker, Nancy Pelosi, or his tweet that “Civil war is inevitable” following riots that erupted recently in the UK.

As the BBC reported, the riots occurred after the fatal stabbing of three girls in Southport. “The subsequent unrest in towns and cities across England and in parts of Northern Ireland has been fueled by misinformation online, the far-right and anti-immigration sentiment.”

Nor does Musk bother with niceties when people disagree with him. Late last year, advertisers considered boycotting X because they believed some of Musk’s posts were anti-Semitic. He told them during a live interview to “Go fuck yourself.”

He has welcomed Donald Trump, the Republican Party’s presidential nominee, back onto X after Trump’s account was frozen over his comments surrounding the January 6 2021 attack on the capitol. Since then both men have floated the idea of governing together if Trump wins a second term.

Is the world better off with tech bros like Musk who demand unlimited freedom and assert their influence brazenly, or old-style media moguls who spin fine-sounding rhetoric about freedom of the press and exert influence under the cover of journalism?

That’s a question for our times that we should probably begin grappling with.

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Even news outlets perceived as politically neutral can benefit from today’s polarized media environment.

The past decade has seen a surge in partisan online news sites that patently skew toward one side of the American political spectrum, from the right-leaning Blaze Media to the openly liberal Axios. Money is a primary driver of this media polarization: User engagement translates to dollars in subscriptions or advertising. Online news outlets’ revenue depends upon attracting eyeballs from an increasingly polarized reading public known to avoid journalism that does not confirm their existing biases.

Surprisingly, those same dynamics can also attract readers of all stripes to officially nonpartisan outlets. That’s the unexpected takeaway of a paper I co-authored and published in the June 2024 issue of the Journal of Management Information Systems. It found that news sources perceived as unbiased can sometimes tweak their story presentation to appeal to partisans.

That is good for their bottom line, at least in the short run—but it may not be great for American democracy.

Polarized politics, polarized readers

My co-authors and I studied the interplay of political partisanship and competition for audience attention between three hypothetical news outlets: one left-wing, one right-wing and one centrist.

In a series of studies using computer models and game theory, we simulated news outlet competition in different types of media markets, including one with minimal regulation, such as the United States.

Research shows that as American readers have grown more partisan over the past 30 years, so have many American news sources. At the same time, a handful of websites perceived as neutral, such as The Hill and Reuters, have gained prominence without much controversy.

We wanted to know: How might the U.S. “marketplace” of information with increased polarization evolve?

Our model assumed that, in a market with minimal regulation, competition among news media firms is a main driver of change.

Surprisingly, we found that news outlets perceived as relatively neutral may have a competitive advantage over partisan media on both sides because they can still shift rightward or leftward. Unlike publications with an already well-articulated political niche, centrist media retains the flexibility to slant.

This partisan tilt could begin with small, isolated wording changes. But if those tweaks draw increased engagement from partisans, neutral news websites would be encouraged to advance toward a more permanent polarization.

Essentially, they’d test the boundaries, seeing how far to the left or right they could tack without losing their perceived neutrality.

Has this process already begun?

Many avowedly neutral websites have recently been accused of becoming more partisan, a charge they deny.

The Associated Press, a venerable American wire service, was criticized in July 2024 for allegedly using language favoring President Joe Biden over former President Donald Trump. In 2019, the owner of the politics website The Hill, Trump associate James Finkelstein, came under fire for the website’s “conspiratorial” coverage of Biden’s dealings in Ukraine.

Still, a few doggedly neutral online outlets remain.

The website AllSides categorizes news stories as “from the left,” “from the right” or “from the center.” It names about 10 news sources widely read by Americans as “unbiased.” Among them are Newsweek magazine, the U.K.-based news agency Reuters and The Hill, which recovered its centrist footing after it was sold by Finkelstein in 2021.

Our research indicates that these doggedly centrist outlets could attract a larger, more partisan readership by shifting their presentation style slightly to the right or left.

Initially, that might be a good thing: Reading more centrist news would get these readers out of their partisan echo chambers. However, our findings suggest that, over time, engagement from partisans would encourage previously neutral outlets to pander to said partisans.

Polarization sells better than neutrality, but at the cost of journalistic objectivity.

Democracy depends on the press

The loss of unbiased media would not be good for American democracy.

Democracies need an informed electorate that makes political decisions based on fact, not dogma. The media plays an essential role in this task, as U.S. law once recognized. The 1949 fairness doctrine of the Federal Communications Commission required broadcast networks to give airtime to differing opinions on public matters.

The agency formally repealed the doctrine in 1987, citing First Amendment concerns. Similar constitutional issues would likely prevent the revival of the fairness doctrine and stymie the creation of any similar constraints on digital and print media today.

Barring direct intervention from regulatory authorities, neutral news outlets have an economic incentive to win the battle of the online press echo chambers. Seizing it would eliminate one remaining forum for fact-based discourse.

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Can artificial intelligence be an effective wingman?

Roman Khaves, co-founder and CEO of Rizz, thinks so. The AI dating assistant app analyzes screenshots of conversations you’re having on other platforms—from Tinder to iMessage—and crafts reply suggestions. Rizz has amassed millions of downloads since launching in 2022, with more and more competition also emerging into the dating scene since.

Khaves recently spoke with The Associated Press about the demand for this kind of platform, how AI is being used to find in-person connections, and what guardrails are needed to bring AI technology into such an intimate part of our lives. The interview has been edited for clarity and length.

Q: How does Rizz work?

A: Rizz is an AI dating assistant. If you’re stuck in a conversation or you don’t know how to open up on a dating app, you can pull out Rizz and upload a screenshot of a particular profile or conversation that you’re having trouble with on that other platform. It will give you tons of suggested replies—and also a strategy on how you could start the conversation or respond to certain things.

The more you use Rizz, the more it understands what type of replies you like. And that trains the model to improve over time.

Q: Why launch this kind of app?

A: Three years ago, I personally found myself stuck in the middle of dating app conversations. Sometimes, I would take a screenshot to send to some of my friends and say, “Hey, what do I say next?” And we knew there had to be a better way.

The moment that ChatGPT hit the scene, that’s when we launched Rizz. We integrated the API technology with our product to make it possible. And we saw immediate traction. We’ve had over 7.5 million downloads to date, growing about 30% month over month, since launching in 2022. There’s clear demand for this type of product.

Q: Who are your biggest competitors?

A: I’d say at least 40% of our screenshots and conversations come from messaging apps like iMessage and Instagram. We get a lot of Instagram, which I would argue is the biggest dating app in the world today—and Instagram recently added an AI feature that can give you different variations of how to rephrase a message, like more funny or witty suggestions.

So, for now, Instagram is a huge competitor. Tinder also has suggested openers. It’s still very different than what we do—we take it a step further and we give users a strategy—but I think those are the biggest competitors, as well as ChatGPT. And there’s many similar apps now and new ones emerging every single day.

Q: What would you say to folks who may be concerned about bringing AI into their dating life? Where do there need to be guardrails?

A: We’re not trying to replace human connection. We’re just trying to help people get to the date in the first place—because it is hard out there. A lot of people don’t even know how to start conversations online, and I think AI is helping boost that confidence.

Still, privacy and control is critical in this new world of AI. We could have built Rizz very differently. We could have made it so that Rizz would respond automatically or connect to the dating apps—but we don’t do any of that. All users are in control of exactly what gets sent out to their match. And we are going to continue building with these priorities in mind.

Q: What’s next for AI and dating?

A: A lot of the big dating companies are thinking about how to leverage AI in order to enhance human connection in this space today. That’s also what we set out to do, and are still improving on.

Rizz is currently built to help you get off dating apps as quickly as possible. Next, we might want to help you beyond the first date, perhaps with future date ideas or something like birthday gift ideas. It’s really about supporting the user throughout their journey.

© 2024 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed without permission.

Rating agency Fitch recently warned that the rapid spread of the mpox virus in sub-Saharan Africa could add to the fiscal pressures many countries in the region are already experiencing.

The Africa Centres for Disease Control and Prevention and the World Health Organization have declared the latest outbreak of mpox in Africa a health emergency. An epidemic in the Democratic Republic of Congo has spread to neighboring countries.

Seven countries rated by Fitch—Cameroon, Côte d’Ivoire, Kenya, Nigeria, Rwanda, South Africa and Uganda—have confirmed mpox cases.

Fitch cautioned investors about possible under-reporting of mpox cases and that the outbreak could accelerate, raising the prospect of increased pressure on government finances.

But is this alarm call necessary? Or is it exaggerated?

Based on my research into rating agencies over the past 10 years, there are clear biases in the way they determine African sovereign risk. Fitch’s statement can be viewed as another case of a rating agency looking at events in Africa through a more negative prism than the one it uses for countries in the west.

Several studies have found evidence that there are biases with rating agencies overstating certain risk factors on the continent.

A comparative analysis of 30 countries in Africa and other regions highlights a lack of uniformity in the application of the economic indicators in ratings. This lies behind the African Union’s decision to adopt a declaration on the establishment of an Africa Credit Rating Agency.

But some rating analysts have come to the defense of rating agencies, arguing that there is no bias against African countries.

For their part, rating agencies maintain that their methodologies are objective. And a recent article by news agency Reuters claims that there were no studies presenting evidence of statistical bias in ratings against Africa.

In my view these claims raise the question: what measure is being used to assess bias? This is important because bias can manifest in different ways—through decisions about what to measure (quantitative), or through more subtle forms of qualitative bias.

I argue in this article that credit ratings are biased against Africa in subjective ways. And that one of the key contributing factors is the location of rating analysts.

A thin presence

Most rating analysts are based in Europe, Asia and the US. Of the big three, Standard & Poor’s and Moody’s each have a single office in South Africa. They have a total of five to 10 analysts covering about 25 sovereigns, corporates and sub-sovereigns. Fitch Ratings closed its only Africa office in 2015.

This raises questions about the workload of analysts and the accuracy of their ratings.

Rating analysts based abroad visit the countries they rate for a maximum of two weeks in a year.

This is insufficient time for analysts to adequately understand and evaluate risk factors. Inadequate consultations and short visits have led to analysts basing their assessments on pessimistic assumptions, desktop reviews, virtual discussions and publicly available information.

These processes have also omitted critical data that often is best obtained by being inside a country. Estimations of subjective risk factors in policy effectiveness, quality of insitutions, political and geopolitical dynamics. The conservatism of analysts, a lack of understanding of the context and ratings errors have been a recurring feature of African ratings.

Research shows that familiarity with a country, being closer to a country being rated and home bias (towards the home country of a rating analyst) result in analysts assigning better sovereign rating scores than they do to countries they are not familiar with or live very far from.

Where there’s room for bias

To understand how bias might happen, it’s important to break down the credit rating methodology.

For example, S&P Global’s sovereign rating methodology looks at five key factors. Two are primarily quantitative—economic and monetary. The other three are essentially qualitative—institutional, external and fiscal.

On qualitative factors, rating agencies use a score from 1 to 6. Rating analysts have considerable discretion in assigning qualitative judgments on the scores. The judgments could easily be driven by bias.

Credit rating researchers Patrycja Klusak, Yurtsev Uymaz and Rasha Alsakka have found that a connection between a European finance minister and a top executive at one of the three international rating agencies can favorably skew a rating decision.

A finance minister’s connection to a rating agency’s director, executive or senior analyst could raise a sovereign rating by between 0.5 and 1.3 notches.

African rating errors

Here are some examples of errors made by rating agencies in their assessment of African countries.

Tunisia: Fitch made an error in a December 2022 review of Tunisia. Fitch published its rating of Tunisia outside the scheduled calendar, without considering all available and relevant information. Fitch corrected this three months later, but only to comply with the European Securities Markets Authority’s regulatory requirement for rating agencies not to deviate from the calendar of sovereign rating publications. In my view, this error could have been avoided if Fitch had had a local presence in the country.

Cameroon: In affirming Cameroon’s Caa1 rating, Moody’s viewed the government’s decision to grant a 5% salary increase to civil servants and other public sector workers as negative. Cameroon also suspended a proposed additional personal income tax, which Moody’s interpreted as a negative development for the country’s credit profile. But Cameroon fiscal metrics were moderate. The country could afford the pay hike. Markets considered its fiscal budget as modest relative to its peers with a similar population profile.

In addition, rating agencies had wrong ratings for Cameroon from November 2022 to August 2023. Their ratings were based on unverified information on late external debt service payments between January and November 2022. The rating agencies’ action one year after, and reports on Cameroon’s delayed debt service payments, suggested that the agencies did not have timely and factual official information. If the agencies had had sufficient engagements with the relevant government officials, their analysts would have had such information.

Nigeria: Moody’s had to reverse its downgrade of Nigeria’s outlook within seven months. As the reason for changing its mind, it cited positive economic policy developments from the government’s removal of fuel subsidies and the country’s unified foreign exchange rates. But these economic factors had not changed between the downgrade and its reversal. The Nigerian government had challenged Moody’s initial decision, arguing that the rating agency didn’t understand the country’s domestic environment.

The reversal of Nigeria’s rating direction in the short term could be evidence that the rating agency had erred in its initial analysis.

Way forward

Locating more analysts on the continent and widening the scope and time of consultations will address some of the biases in how rating agencies assess African countries. Where analysts are already Africa-based, they need to extend their scope of stakeholder consultations and to spend more time in the countries they rate.

The interpretation of events and perception of risk by locally based analysts will be different from those who are foreign based. This will partly address contestation around the bias and lack of adequate consultations in Africa ratings.

This article is republished from The Conversation under a Creative Commons license. Read the original article.