National

From the Valley to D.C., Big Tech players like Facebook, Google and Amazon are under more scrutiny than ever as new technology develops and privacy and antitrust concerns grow in lockstep with companies’ ambitions.

Technology

Players have increasingly bristled over talkative characters, just as studios are betting on buzzy AI tech to add even more dialogue to games.

Driving the news: Developers of Forspoken, one of 2023’s earliest big releases for PlayStation 5, updated the game this week to adjust how chatty the main character, Fey, and her magical companion can get.

The action-adventure game is partly known for its quippy banter, both in storytelling cutscenes and amid gameplay.

Tuesday’s patch notes say that the tweak “adjusted the frequency and content of the banter between Frey and Cuff.”

The big picture: The Forspoken changes follow growing irritation among some players over the quantity of dialogue. As game content has boomed, so has the number of spoken lines in a typical triple-A release.

Last year’s Dying Light 2 was lampooned for boasting about its completion time of 500+ hours, and likewise caught flak for touting “350,000 words, 40,000 lines of dialogue.”

2022’s Horizon Forbidden West and God of War Ragnarök have also been dinged in coverage for extensive chatter.

“As I traversed through Horizon Forbidden West’s beautiful post-post-apocalyptic America as Aloy, I was bombarded with her constant soliloquy on where to go and what to see,” Jordan Sutlive wrote for Polygon.

Redfall, Dead Island 2 and Atomic Heart are other recent games that have made players pine for the days of silent protagonists.

Between the lines: It’s not only the sheer amount of dialogue that has drawn criticism, but also modern usage in game design and its tone.

“Unfortunately, I think a lot of folks have confused ‘a funny character’ with ‘a character who doesn't take their own situation, no matter how perilous or interesting it is, seriously,’” Austin Walker, IP director at game studio Possibility Space and former editor in chief of Vice’s Waypoint, wrote in a blog post about the recent explosion of irreverent, Marvel-esque dialogue writing in games.

Meanwhile, dialogue has been increasingly employed to aid the player — whether the player wants that aid or not. Multiple outlets have informed players you cannot disable God of War Ragnarök’s verbose puzzle hints.

The intrigue: Player irritation is mounting just as studios are experimenting with ways to use AI to give more characters even more to say.

Assassin’s Creed developer Ubisoft unveiled its generative AI Ghostwriter tool in March during the Game Developer Conference. It’s designed to let a writer come up with a character and situation, then generate numerous lines that character might say.

Meanwhile, the use of AI to voice characters is already being hyped by some tech companies as a way to ease the addition of dialogue to games.

Reality check: Forspoken’s recent tweak follows a bumpy release.

The game faced a round of viral tweets mocking its dialogue before catching middling reviews for its flat story and repetitive gameplay.

Even before the most recent update, Forspoken had a slider in its settings to let players adjust the volume of banter.

Yes, but: Both voice acting and writing can be fraught professions in gaming.

Actors can face uphill battles in fighting for compensation, while writers often face development crunch.

What’s next: If you’re turned off by the chitchat, enjoy Link’s stoicism in Tears of the Kingdom and prepare yourself for the next epic RPG: September’s Starfield.

Meanwhile, Forspoken’s In Tanta We Trust DLC comes out Friday.

Sign up for the Axios Gaming newsletter here.

Players detest chatty video game characters, but here comes AI

Forspoken, God of War Ragnorök and Horizon Forbidden West have been called out for verbose characters.

Video games

Gaming

Communicators

Axios Communicators

Rebecca Hahn has joined Tools for Humanity — the parent organization of WorldCoin, co-founded by Sam Altman, Alex Blania and Max Novendstern — as chief communications officer, Axios has learned.

Why it matters: Hahn was one of the last communications professionals standing at Twitter under Elon Musk, and is now working for OpenAI founder and Musk rival Altman. 

Before Twitter, Hahn served as chief communications officer for the electric mobility company Bird and was a partner at The Outcast Agency. 

Details: Since quietly joining Tools for Humanity this year, Hahn has been slowly building the organization's communications apparatus by hiring Twitter and Bird alum John Wachunas to oversee editorial content and enlisting outside agency support from Riff City Strategies.

Driving the news: This announcement comes on the heels of a $115 million fundraising round led by Blockchain Capital, with investments from a16z Crypto, Bain Capital Crypto, and Distributed Global.

The big picture: While AI has proved to be a lucrative business, winning over the trust of the general public is an uphill battle for communicators.

This year's Axios Harris Poll 100 brand reputation survey found that 75% of American adults are concerned about deep fakes and other generative AI content.

And it's unclear how customers will take to these biometric recognition technologies that are meant to protect users from the generative AI tools that Altman made so easily available. 

What they're saying: "My passion is working with technology and teams with the potential to have a transformative impact on humanity and the planet," Hahn told Axios. "Tools for Humanity and its support of the Worldcoin project are a perfect complement to this passion."

Scoop: Twitter alum Rebecca Hahn to lead communications for Sam Altman's Tools for Humanity

Hahn was the last communications professional standing at Twitter under Musk and is now working for OpenAI and Musk-rival Sam Altman.

Communications

Tech Policy

One more concern to add to the long list of fears stoked by the rise of generative AI: Experts say it could incite more antisemitism in the U.S. at an already fraught time for American Jews and other groups targeted by hate.

Driving the news: Generative AI chatbots like OpenAI's ChatGPT and Google's Bard, which respond to prompts with written responses pulled from web data, have sparked a frenzy over the capabilities and potential dangers of AI's rapid technological advancement. 

How it works: Because AI models learn to complete sentences by analyzing enormous quantities of text created by people, usually on the internet, they pick up bias embedded in both the digital environment and broader society.

Antisemitic incidents reached a new high in the U.S. in 2022, the Anti-Defamation League said in a March report, citing a 36% rise from 2021. That includes harassment, vandalism and assaults.

What they're saying: "We are very concerned about how these models are being created," Anti-Defamation League CEO Jonathan Greenblatt told Axios' Russell Contreras.

Greenblatt pointed to how quickly Microsoft's Tay, a 2016 AI chatbot, was shut down after it spewed a series of lewd and pro-Nazi tweets soon after launch — a stance it borrowed from other Twitter users.

He said while he's appreciative of some initial improvements made to ChatGPT, tech companies need to be open and transparent about data sets and algorithms. Calls for transparency are growing; mostly, companies don't disclose what data is being used to train these models.

The big picture: Experts say that the dynamic of AI-amplified antisemitism can be one good bellwether or proxy for assessing the much wider varieties of bias and hate that the technology can spread on the basis of race, gender or LGBTQ identification, religion, immigration status, and other factors.

By the numbers: An ADL survey of 1,007 adults in the U.S. found that 84%  are worried generative AI tools will increase the spread of misleading or false information.

An April report from the Center for Countering Digital Hate found that when prompted about hot-button topics around hate, misinformation and conspiracies, Google's Bard chatbot produced text with misinformation 78 out of 100 times. 

Google said at the time Bard is an "early experiment" prone to "sometimes [giving] inaccurate or inappropriate information."

The intrigue: CyberWell, an Israeli nonprofit platform that monitors social media antisemitism in real time, is working on "high-integrity data sets" identifying online antisemitism based on keywords and other specifics that can be used in the future to train generative AI and weed such content out, CEO Tal-Or Cohen Montemayor told Axios. 

"I think people have understood since the rollout of social media, the social harms that are caused and the direct connection we see between online hate and hate crimes in real life," Cohen Montemayor said. "We understand we want a more ethical result when it comes to generative AI."

A March report from CyberWell showed wide disparities in how current major social media platforms remove such content, which is against  the company's terms of service but nonetheless proliferates online.

"Antisemitism is one of the most nuanced and layered forms of hate speech out there because of the history and modern manifestations of it," Cohen said. "But it is one of the best forms of hate speech to start training generative AI on, because it's so nuanced."

Go deeper: Antisemitism is especially hard to tackle in generative AI because it takes many forms. It can be a picture, a phrase, a veiled generalization, basic misinformation or derogatory language.

Generative AI programs are "trained on masses and masses of publicly available material on the internet, including social media content," Callum Hood, head of research for the Center for Countering Digital Hate, told Axios. 

"You could knock out websites like Stormfront and so on, but we know that antisemitic conspiracy theories occasionally rear their ugly head in mainstream media."

He said: "We know the tech companies that have ingested this stuff to train their AIs have not done the best job of cleaning misinformation out of the training material ... and then that stuff ends up in the answers."

Most AI developers right now say they are intervening to try to root out bias — both by removing it from training data sets up front, and by adding rules and guardrails to the chatbot later.

But some companies and leaders in the field — including OpenAI co-founder Elon Musk, who has since broken with the company — criticize these efforts as being too "woke" and prefer what they claim is a "free speech" approach.

Given this political disagreement within the industry, even if big providers try hard to root out bias and hate speech, at least some widely available AI programs are likely to end up promoting it.

How generative AI could generate more antisemitism

Hate and hate speech are tough to root out when AIs are trained on vast troves of online human speech.

Illustration of a star of David being glitched and distorted. 

Artificial intelligence

Automation and AI

A Chinese state-sponsored group has hacked into critical American infrastructure, including in the U.S. territory of Guam, Microsoft and the "Five Eyes" intelligence alliance warn.

Why it matters: Guam is home to three American military bases. The western Pacific island would play an important strategic role should the U.S. need to respond to any potential Chinese military attack on or blockade of Taiwan.

The likely aim of the operation is to "disrupt critical communications infrastructure between the United States and Asia region during future crises," per a blog post Wednesday by Microsoft, which detected the hacking.

Driving the news: The U.S. Cybersecurity and Infrastructure Security Agency (CISA) issued a joint advisory with its "Five Eyes" partner agencies in the U.K., Canada, Australia and New Zealand warning that the "Volt Typhoon" hackers posed a threat to all five allied countries following a "recently discovered cluster of activity."

Microsoft said in its blog post that the "state-sponsored actor based in China ... typically focuses on espionage and information gathering."

Detecting "and mitigating this attack could be challenging," Microsoft said.

State of play: "Volt Typhoon has been active since mid-2021 and has targeted critical infrastructure organizations in Guam and elsewhere in the United States," per Microsoft.

"In this campaign, the affected organizations span the communications, manufacturing, utility, transportation, construction, maritime, government, information technology, and education sectors. Observed behavior suggests that the threat actor intends to perform espionage and maintain access without being detected for as long as possible."

The Australian Signals Directorate’s Australian Cyber Security Centre noted that a key strategy of Volt Typhoon, known as "living off the land," used "built-in network administration tools to perform their objectives," allowing the group "to evade detection by blending in with normal Windows system and network activities."

The big picture: The Chinese Communist Party has been linked to previous hack attacks targeting the U.S. government, businesses and American infrastructure, but it has always denied being involved in such cyber espionage.

What they're saying: "Today’s advisory highlights China's continued use of sophisticated means to target our nation's critical infrastructure, and it gives network defenders important insights into how to detect and mitigate this malicious activity," CISA director Jen Easterly said in a statement.

Chinese hackers hit critical U.S. infrastructure, intelligence agencies warn

The likely aim is to "disrupt critical communications infrastructure" between U.S. and Asia in a crises, Microsoft says.

The aircraft carrier USS Theodore Roosevelt, is docked at Naval Base Guam in Apra Harbor amid the coronavirus pandemic on Monday, April 27, 2020. 

The aircraft carrier USS Theodore Roosevelt, is docked at Naval Base Guam in Apra Harbor amid the coronavirus pandemic on April 27, 2020. Photo: Tony Azios/AFP via Getty Images

Computer hacking

China

China is becoming the most formidable rival America has faced, and the risk of confrontation with Russia, North Korea or Iran still looms. Authoritarians are rising, the chaotic Afghanistan exit revealed the limits of U.S. power, and the pandemic, climate change and other critical challenges continue to pose grave threats.

World

President Biden has made the COVID-19 crisis and a post-Trump return to national unity and traditional democratic ideals his top priorities. From vaccinations to stimulus to schools, Biden is seeking bipartisan compromise while showing a willingness to use executive authority and bare Democratic majorities in the U.S. House and Senate to implement his policies. Republican leaders are navigating deep party divisions over if and how to move beyond former President Trump.

Politics & Policy

After a record economic expansion, the coronavirus pandemic sent the U.S. economy into a recession as unemployment soared to staggering heights. The country now faces urgent questions about how much stimulus is needed for reeling consumers and businesses, and what a recovery might look like.

Economy & Business

Axios Gaming

Sony plans to release a new handheld later this year that will play games streamed from the PlayStation 5, a partial return to portable gaming after a decade away.

Why it matters: The device, codenamed Project Q, will further test the potential of cloud-based gaming, just as it'll test whether Sony can rediscover its footing in a market where it struggled.

Details: PlayStation chief Jim Ryan introduced the device during an hour-long showcase of upcoming PlayStation games, saying it will support streaming over Wi-Fi.

The device, which resembles a phone screen slotted between halves of a PS5 controller, has an 8-inch screen.

Yes, but: Sony's last foray into portable gaming, the powerful PlayStation Vita, failed to outsell the weaker Nintendo 3DS, and the handheld gaming market has gone on to be dominated by mobile games and Nintendo's hybrid Switch console.

A streaming-only handheld device from Sony is a big risk.

Most of the PlayStation showcase focused on game reveals. The biggest ones:

Destiny-maker Bungie, bought by Sony last year, confirmed that it is making a new version of Marathon, the shooter it released for the Mac in 1994. Slated for PS5, Xbox and PC, it'll focus on player-vs-player and won't have a single player campaign.

Konami is remaking the classic PlayStation 2-era military espionage game Metal Gear Solid 3: Snake Eater. It revives a franchise Konami has largely left alone since series creator Hideo Kojima's acrimonious split with the publisher in 2015.

Sony-owned Haven Studios showed a cinematic, non-gameplay trailer to reveal Fairgame$, a competitive heist game about robbing the ultra-rich.

Co-creators of indie hit Night in the Woods are making a new game called Revenant Hill.

Early PS4 multiplayer game Helldivers is getting a sequel for PS5 and PC.

Artful indie studio Giant Squid (Abzu, The Pathless) is making Sword of the Sea, a game that includes surfing across sand dunes.

SquareEnix is developing a multiplayer shooter called Foamstars that appears to be like Nintendo's popular paint-shooter Splatoon, but with foam.

Be smart: Many of the games announced by Sony today are for 2024 and beyond. 

Sony's main PlayStation Studios game release for this year appears to be Spider-Man 2, which closed the event with a new trailer.

Several of Sony's announced projects, including Bungie's Marathon, Haven's Fairgame$ and Firesprite's Concord are focused on multiplayer, a category where Sony has said it needs to improve. 

PlayStation to release a new streaming handheld Project Q

It was announced during a showcase filled with multiplayer game reveals and a Metal Gear return.

Screenshot: Sony Interactive Entertainment, via YouTube.

Crypto

Axios Crypto

A lab-made opioid makes its way as chemical components from China to a shop in Latin America, where it's made into fentanyl, packaged and trafficked into the U.S. — a drug move that crypto can track.

Why it matters: Criminals use crypto as a means to operate — but that also means that law enforcement can leverage blockchain analysis to stop them.

Driving the news: A report by blockchain analysis firm Chainalysis traced the flows of fentanyl described by a Drug Enforcement Administration intelligence report, using crypto transactions involving suspected chemical shop addresses.

Zoom in: "We weren't mapping out the whole ecosystem," Eric Jardine, cybercrimes research lead at Chainalysis, told Axios. "It was a look at the way in which crypto can facilitate the trade in these chemicals."

Indeed, crypto research firm Elliptic on Tuesday put out its own investigative report showing that the vast majority of chemical shops accept payment in bitcoin and tether.

Details: China was identified by the DEA as the primary source of fentanyl and fentanyl-related substances ultimately trafficked into the U.S. in 2019, with Mexican transnational criminal organizations producing the synthetic opioids before smuggling them across the U.S.-Mexico border.

The cryptocurrency addresses linked to known China-based chemical shops, as well as others that Chainalysis identified, received some $38 million in crypto from Jan. 2018 to April 2023, according to the firm's report.

Meanwhile, fentanyl is also sold on darknet markets that ban those synthetic opioids for fear of attracting the attention of U.S. law enforcement. Some vendors use the street names of the products like "China White" to circumvent scrutiny.

The big picture: Deaths by drug overdose in the U.S. have climbed in recent years, with those involving synthetic opioids like fentanyl rising in-step.

Fentanyl is involved in more deaths of Americans under 50 than any other cause of death. Meanwhile, states are wrestling with the crisis — candy-colored versions were recently spotted in Ohio.

The bottom line: Proper blockchain analysis can enable law enforcement to monitor fentanyl sales, identify criminals, and provide crucial evidence to prosecutors, Chainalysis wrote. Ultimately, crypto could play a positive role in reducing drug trafficking into the U.S.

Crypto could be tool in fight against drug trafficking

Criminals use crypto as a means to operate — that means blockchain analysis can be leveraged to stop them.

Illustration of a pill casting a shadow, with the pill made from dotted lines as if it were invisible. 

Cryptocurrencies

The sprawling Swedish video game company The Embracer Group is reeling following its announcement that a "transformative" partnership that's been negotiated for months fell apart Tuesday night. 

Driving the news: Embracer said the collapse of the mega-deal, the partner of which still has not been revealed, was unexpected. Embracer said it had a verbal commitment in October and described negotiations as "taking far longer than expected."

Shares of Embracer Group were down more than 40% this morning.

Details: "All documentation was finalized and ready to go as of yesterday," Embracer CEO Lars Wingefors told investors Wednesday morning. "We asked for the execution of the agreement before our Q4 announcement. However late last night we received a negative outcome from the counterparty."

Compounding the bad news, the company lowered its earnings forecast for the next year, citing game delays, from a range of $965 million - $1.3 billion down to $655 million - $840 million.

Catch up fast: Embracer has been a voracious purchaser of games, game studios and intellectual property, making more than 20 acquisition deals last year, valued at over $1 billion, according to a Drake Star report.

Embracer's portfolio includes rights to "The Lord of the Rings" and both the characters and studios tied to the Tomb Raider franchise.

The company also accepted a $1 billion investment from the Saudi-government-funded Savvy Games Group last year.

What they're saying: "It's been a rough night," Wingefors said in a livestream today.

Asked if the negative outcome was the last word on the partnership, Wingefors said the message Embracer got from the would-be partner was that the deal "made perfect sense, so it was other decisions impacting them than the agreement itself."

$2 billion partnership deal falls through for Embracer Group

The Swedish owner of Tomb Raider, Lord of the Rings, sees stock plummet 40%.

Video game screenshot of zombies on a beach

The Embracer Group's Dead Island 2, which sold more than 2 million copies since its April launch. Image: Embracer Group.

Axios AI+

Change in 2023 Axios Harris 100 reputation scores among select tech companies

Data: Axios Harris poll; Chart: Alice Feng/Axios

Large tech companies crowd both ends of the rankings in this year's Axios Harris Poll 100 brand reputation survey, with those that provide tangible products and useful services clustered near the top and those that trade in social-media messages banished to the bottom. 

What's happening: Worries over the rise of AI, on top of long-building distrust of social media providers, have pushed the public toward these extremes. 

75% of surveyed American adults said they are concerned about deep fakes and other generative AI content.

The highest-scoring tech companies are Samsung, Amazon and Apple, all in the retail and hardware game. 

TikTok, Meta, and Twitter are locked in a race with Fox, the Trump Organization and failed crypto exchange FTX for the worst reputation among the top 100 brands.

Winners: The tech companies with the highest Reputational Quotient, or RQ®, scores, were Samsung (81), Amazon (80.7) and Apple (80.6), which retained their 2022 position. Sony (79.8), Microsoft (79.7), and LG (78.8) also  landed in the top 20. 

Millennials ranked Google highest, though the company ranked only 35th overall.

JP Morgan Chase beat out Apple and Paypal to receive the highest data privacy protection score.

Made in the U.S.A. brands soared in 2023, with third-ranked John Deere leaning into “high-tech field management technologies, such as a robotic-based fertilizer system and a new electric excavator,” per Harris, to harvest a better reputation.

Losers: No social media platform ranked higher than 94th.  

Tiktok, with an RQ of 61, ranked 94th, followed by Facebook (Meta) in 97th and Twitter in 98th. All received lower RQ scores in 2023 compared to 2022. 

Meta was the company least trusted by women (only 21% gave the company a top trust rating), and the second-least trusted by Gen Z (16%), ahead of only the Trump Organization. 

Yes, but: Meta is now the best performing stock of the year — more than doubling in value since January — suggesting that investors are more cheered by cost-cutting measures than worried about the company's reputation. 

Between the lines: Elon Musk’s chaotic takeover of Twitter hurt the social platform, and showed Tesla stakeholders weaknesses in Musk's manic leadership style. 

Tesla's reputation dropped 50 places from 12th in 2022 to 62nd place this year. No other company dropped more than 22 places. 

Meanwhile, a generational split is leaving TikTok exposed, despite its large user base — paving the way for more bans on the service. 

The generational split over Tiktok's reputation is the biggest of any of the companies on the Axios Harris 100 list.

Boomers, who occupy roughly half the seats in Congress, gave TikTok a RQ score of just 41, compared to 68.7 for Gen Z / Millennials. 

Even the Gen Z and Millennial respondents who make up TikTok's user core rank the company only 87th.

Layoffs are priced in as a cost of American business life and left firms with lo lasting reputation damage.

No big tech company engaged in mass layoffs in recent months took a reputation hit traceable to the job cuts, and most maintained their RQ scores. 

Ryan's thought bubble: People can see and somehow sense more tangible benefits from the Apples and Amazons and Samsungs.

They employ more people because they make and sell things.

You can see and hold those things so you process them differently from digital media.

In most cases, you paid them money for what they offer, instead of paying in data (though sometimes it's both).

They're less often drawn into the political controversies that divide social media users, whereas the big social platforms stoke those arguments to inflate engagement and boost profits. 

Ina's thought bubble: Social media companies are likely always going to have a tougher task than hardware makers and retailers for all the reason Ryan lists.

However, that doesn't mean the actions taken by companies in either group can't have a big impact on how they perform relative to their history and their peers.

It wasn't that long ago (think pre-Cambridge Analytica) that Facebook and Twitter were held in much higher esteem.

I'll be interested to see how companies such as OpenAI fare in future surveys.

Go deeper: Full results and poll methodology

long-building distrust of social media providers

Axios Harris 100: Americans still trust tech firms that deal in goods

But social media firms have continued a long reputational dive.

Thursday's technology stories