Top Stories
AI Briefing: How AI showed up at Cannes Lions 2024

Top Stories Tamfitronics

By Marty SwantJune 24, 2024

Ivy Liu

Digiday covers the latest from marketing and media at the annual Cannes Lions International Festival of Creativity. More from the series →

At this year’s Cannes Lions Festival of Creativity, it’s no surprise AI was the topic of the weekwith cocktail conversations along the Croisette that ranged in strength from hype as potent as absinthe to the more modest hypothetical rosé. But there was enough buzz to make even the most sober points seem like chatbot-worthy hallucinations.

For the second year in a row, generative AI was a main theme at Cannes Lions 2024. Conversations have matured since Cannes Lions 2023 — from early questions about AI’s role in marketing to newer explorations of practical applications — but advertising execs still noticed an excess of AI jargon.

Some saw AI news at Cannes Lions 2024 as focused on future AI commitments. Others note marketers privately admit to being overwhelmed by options and also still underwhelmed by results. There are also outstanding questions like who should own AI inside a company — and who owns the budget.

The experience layer with generative AI will be more profound for marketers than the product layer, said Huge chief brand officer Matt Weiss. He thinks “intelligent experiences” (IX) will change everything for how consumers interact with platforms.

“Everyone knows the words and rhetoric,” Weiss said. “They know what to say. But do they know what to do? Not so much. Not possible. The landscape is changing too rapidly.”

With so many AI offerings seeming similar, differentiation is keynoted Elav Horwitz, who heads up innovation at IPG’s McCann Worldgroup. That makes creative and unique applications especially important if companies want their AI offerings to stand out. Horwitz mentioned AI firms should take the ad industry’s playbook and find ways to communicate what’s most compelling about their products.

This year, tech firms brought with them “AI labs” to show potential customers what might be possible with generative AI, via playful pilots and beta tests. However, Horwitz said some were limited to mostly imagery and visuals that lacked the depth of strategy, insights and concept development.

“AI and tech companies should take a page from the advertising playbook, understanding how to differentiate their offerings and communicate their stories more compellingly,” she said. “This includes emphasizing responsible adoption and their commitments to safety for brands and humans, while also helping companies innovate and solve problems rather than just focusing on productivity.”

An amuse-bouche of AI news from Cannes Lions 2024:

  • Platforms: Meta announced a new chatbot for Messenger, and a new API for content management on Threads. Meanwhile, AI products were announced by Reddit, Yahoo and Spotify. TikTok used the week to debut a way to make ads with AI avatar influencers, while Adobe announced a new partnership to let Adobe Express users create AI-generated content for TikTok using TikTok’s library of commercially available music.
  • Agencies: Havas used the week to unveil a new “Converged” strategy for the agency focused on creativity and personalization that includes a €400 million investment in a new group-wide “operating system.” News from WPP included the launch of a new AI-enabled production studio with Nvidia and a separate B2B-focused platform in partnership with IBM’s Watsonx. Meanwhile, IPG’s new deal with Disney uses AI for live sports advertising and uses Disney’s own AI-powered “Magic Words” platform to identify events and emotions for ad placements.
  • Ad tech and martech: Zeta Global formed a new partnership with Amazon’s Bedrock platform to improve the use of creative AI agents that help marketers have more control and customization with automated content production. Earlier this month, Zeta Global CEO David Steinberg told Digiday that its Intelligent Agent platform already has built more than 300 agents for customers: “The massive utilization of it is effectively build-your-own data scientist or build-your-own data onboarding, or tell me my most valuable and under-utilized audiences,” Steinberg said.
  • AI-powered Cannes Lions: The awards part of Cannes Lions also featured several campaigns created with generative AI. Pedigree’s “Adoptable” campaign from BBDO, which used generative AI to help dogs find new homes, won an Outdoor Grand Prix. Another winner was the French telecom Orange, which worked with Publicis to create AI deep fakes of the French women’s national soccer team during the 2023 FIFA Women’s World Cup to address sexism in soccer.

Other Digiday stories about Cannes, AI or both:

Prompts and Products: Other AI news and announcements

  • OpenAI co-founder Ilya Sutskever launched a new startup called Safe Superintelligence Inc., or SSI.
  • Several top AI players released new AI models. Anthropic released a new AI model for Claude-3.5 Sonnet to rival ChatGPT-4o,Runway ML released its Gen-3 Alpha model that rivals OpenAI’s Sora, and Google Deepmind announced a new video-to-audio model.
  • Several AI companies have recently hired new marketers for their C-suite. Loreal Lynch, former head of campaigns at Stripe, joined Jasper as its first chief marketing officer. HeyGen announced former Asana CMO Dave King as its new chief business officer. Writer AI hired Diego Lomanto as its new CMO from his previous role at the AI startup Ada.
  • Laptops powered with Qualcomm’s new AI-powered Snapdragon chips were officially released for purchase.
  • Meta paused its AI rollout in Europe after Ireland’s data protection agency raised data privacy concerns.
  • A former Snap engineer launched a new AI social network for people and chatbots called Butterfly.
  • Google is reportedly shifting its DeepMind research lab to help with commercial services.

https://digiday.com/?p=548697

More in Media

Top Stories
Best of cartoons, June 25, 2024

Top Stories Tamfitronics

Best of cartoons, June 25, 2024

Were sorry, this feature is currently unavailable. Were working to restore it. Please try again later.

8 Images

1/8

Credit:Dionne Gain

2/8

Credit:John Shakespeare

3/8

Credit:Cathy Wilcox

4/8

Credit:Matt Golding

5/8

Credit:Matt Golding

6/8

Credit:Matt Golding

7/8

Credit:Matt Golding

8/8

Credit:Andrew Dyson

Top Stories
Australia’s media companies seriously complained of blackmail? Give us a break.

Top Stories Tamfitronics

Parliament has rarely been addressed by representatives of such a toxic and malignant industry than when Michael Miller of News Corp, Jeff Howard of Seven West Media, and Mike Sneesby of the Nine Network appeared before the joint committee inquiring into social media last Friday — demanding Facebook and Instagram be banned in Australia and claiming tech giant Meta was blackmailing the Australian media and government.

Blackmail. It’s hard to find a purer example of pot-kettle namecalling than Australia’s biggest media companies accusing anyone of “blackmailing” governments.

For generations, Australia’s media companies — and especially the commercial television networks — have worked by blackmailing governments. It’s their entire MO: any government that ever threatened their commercial interests, or failed to grant their incessant demands for ever more regulatory favours, knew it would be bashed and belted across every news bulletin in the country until it complied.

For generations, the interests of Australian consumers were subordinated to those of the big TV networks, with successive governments not merely preventing any competitors from entering the Australian media but also presiding over the steady destruction of competition — to the point where our media landscape looks more like that of Beijing than any healthy democracy. First, new channels weren’t allowed. Then Pay TV was blocked in Australia, and when it was allowed, it was hobbled. Digital TV was delayed and regulated to the point of absurdity (remember Richard Alston’s risible “datacasting”?)

The only government that ever truly stepped out of line was the Gillard government, when Stephen Conroy dared to demand the industry regulate itself better around editorial standards. The united response from the media was froth-mouthed rage and the effort was killed off.

The gambling industry has its donations and network of politically connected lobbyists. The fossil fuel industry has its revolving-door appointments. The arms industry controls the national security lobby. The pharmacists have their pensioner-based scare campaigns. But only the big media companies can threaten to blackmail governments and know they can deliver on it reliably, on TV, radio, print and online.

Comfortable, complacent and confident they could blackmail any government into compliance with its demands, Australia’s media oligopoly was the 800lb gorilla of vested interests. Until they encountered, in the US tech giants, 8,000 lb gorillas that took their lunch out of their hands and began happily consuming it right in front of them, as Facebook and Google did — or took their customers away by the millions, as Netflix has.

The response from Australia’s media was the usual — run shrieking to government and demand that the competitors be brought under control with more laws literally designed by the media lobby, in the form of the news media bargaining code. After playing nice for a time, Meta has decided it can’t be bothered observing the time-honoured rituals of the Australian media oligopoly. So now Australia’s media companies demand its banning.

To see News Corp, Nine and Seven in Canberra pressing the case for a ban on social media (for kids, for everyone, what’s the difference) is to witness a gang of now elderly standover merchants trying the same thuggish tactics that worked in the good-old bad-old days — and all pretending that somehow they are doing so in Australia’s national interest.

Consider them. Seven West Media is a grubby right-wing outlet for rapists and war criminals, with journalistic standards more appropriate for the National Enquirer than any decent media company.

Nine Entertainment is plagued with alleged sexual predators whose behaviour is systematically covered up while the company’s journalists purport to hold the powerful to account.

And, biggest of all, News Corp: a suppurating tumour on global media that has, in its climate denialism and support for Donald Trump, endangered the planet and helped bring its foremost democracy to the brink of a fascist autocracy, a collection of propaganda outlets that lower standards of civil discourse, divide and polarise communities, and smear anyone working for the public good in any market they defile with their presence.

That any of them dare to claimthey are the victims of blackmail is hilarious. That none of their diminishing number of journalists and commentators, usually so quick to turn their gimlet eyes on double standards and hypocrisy, have bothered to call out the absurdity, is entirely expected. That they’ll go on trying to assail their competitors using the only tool they know how — threatening politicians with retribution if they don’t regulate them away — is entirely predictable.

Meanwhile, their revenues and audiences, in inverse proportion to their protestations, will continue to drain away. Turns out, spending all that time perfecting the art of blackmail and standover came at the expense of actually giving Australians what they wanted.

Do News Corp, Seven and Nine have a valid argument against Meta? Let us know your thoughts by writing to[email protected]. Please include your full name to be considered for publication. We reserve the right to edit for length and clarity.

Top Stories
AI-generated news is unhuman slop. Crikey is banning it

Top Stories Tamfitronics

Crikey’s first-ever edition arrived in inboxes on Valentine’s Day, February 14, 2000. It included a reflection on Lachlan Murdoch from his former philosophy tutor, pseudonymous insider gossip from the former political staffer Christian Kerr, and lists of journalists who’d previously worked in politics. There was a profile of Rupert Murdoch lieutenant Col Allan titled “Pissing in the Sink”.

From our inception, we have been so very Crikey. There are a lot of things that make up Crikeyness, but central is its humanness. Like most people, we want to cut through the bullshit, know how the world and power works, hear the gossip, tell stories, take the piss and have fun. We write the way we wish other people would write, which is the way we talk to each other, for our readers.

In the near quarter-century of dispatches — during which newsletters went from being cool, to cringe, back to cool and is now swinging back towards cringe again —Crikey has remained remarkably consistent because our job didn’t change. We didn’t pivot to iPad or video. We just kept doing the same old Crikey: having people write about people for other people. Fundamentally, this is what we think journalism is about.

But there is a threat to this philosophy. Artificial intelligence, specifically advances in generative AI, has been the defining tech trend of the past 18-odd months. Every industry has been experimenting with how to use the technology, including the media. Predictably, some are already using it to create the news. While some outlets have been caughtpublishing AI-written articleswithout disclosure, others are openly publishingarticlesandartworkmade by a machineeven though it makes most Australians uneasy.

Our official position at Crikey is that this stinks. To understand our revulsion, you need to understand how generative AI, specifically large language models, work.

The first step in creating one of these models is hoovering up a huge amount of data. Most data is ripped from the internet without permission en masse. Companies that do this have ingested lots of intellectual property belonging to people who did not agree to being sucked into someone’s AI-generated, low-quality information and even some illegal and abhorrent images.

The next step is training on the data to understand the relationships between all the words and phrases. For example, if you input “the founder of Crikey is”, the most common response among its dataset is probably something like “Stephen Mayne”, which is true. But AI doesn’t draw on years of close observation of Australian media when it answers this question. It makes a calculation based on all the times other people have continued that sentence to come up with its response.

The point is that AI does not understand what it’s being asked to do. It doesn’t know what words mean. It does not think or care about who’s reading what it’s putting out. It certainly does not consider the different angles of a story and who stands to gain from accepting the spin. It’s mindless regurgitation, like a press gallery journalist repeating a politician’s promise without interrogating it.

There are a lot of other reasons to dislike the AI hype and its industry of boosters. It’s a hugely energy-intensive technology that requires incredible amounts of power to use and train, significantly more than other computing methods to complete the same task like searching the web.

It produces bullshit — not our words, bullshit is a scientific, technical term — when it frequently and confidently tells lies. Even when it tells the truth, its answers have racial and sexual biases.

The technology can’t be separated from its unscrupulous creators. The industry’s original sin was stealing the world’s data without permission. Everything it produces is fruit from the poisonous tree. Companies say you can opt out of having your data collected, but only told us this after they got their fill of information. And sometimes these companies lie about letting you opt out.

These companies have grown rich and fat by selling subscriptions and advertising on the data they stole from us. This digital colonial mindset continues in AI’s users, too. They use it to reproduce our artour photographs, our writing and our facesall without permission and often to our displeasure.

No matter how advanced it gets, AI can never do the work of journalism. It can’t pick up the phone, go somewhere, speak to people, or think of a new angle or problem or solution. That doesn’t mean it can’t be used as a poor substitute. As we’ve already seen in Australiapeople are already using AI to rip off other people’s journalism for their own gain and to compete against real stuff. They steal the eyeballs and dollars required to do the work that they think is valuable enough to take for themselves.

The current state of AI is an industry built around a wasteful, violating technology used to create poor imitations of work, riddled with errors and biases, that threaten the very systems that produce the information from which they seek to extract value.

That’s a good reason to boycott it, but our reason is simpler: AI-generated content is bad to read and see. It’s bland and beige. It’s slop that doesn’t stand up to scrutiny but is a close enough approximation to something real that it can fill up space where people don’t pay attention. It’s something that someone didn’t care enough about to spend time to write or create.

It is the opposite of what Crikey is. We believe that AI can never replicate our essential Crikeyness. A thousand chatbots in a thousand years would never come up with a title as evocative, incisive or beautiful as “Pissing in the Sink”. Using AI to generate Crikey would be an insult to you, the reader, as well as everyone who has made Crikey what it is. We’ve fiddled with AI image generation and the occasional block of text as a proof of concept in the past, and our editorial guidelines already require that any AI-generated content is disclosed. But we want to go a step further.

Crikey will never use AI to create what you read and see from us. We will spend the time writing the drafts, we will make the calls, we will read the documents, we will scheme for ways to piss off the right people. We will do the hard, time-consuming, expensive work — because if we didn’t, we wouldn’t be doing what our audience deserves. We join a movement of people rejecting AI-generated content to fight for the real stuff, and we invite other news organisations to join us.

We promise that everything you read and see in Crikey is made by a real person who gives a shit.

Want human-generated news? Subscribe to our free newsletters. Do you support our policy on artificial intelligence? Let us know your thoughts in the comments or by writing to[email protected]. Please include your full name to be considered for publication. We reserve the right to edit for length and clarity.

0
YOUR CART
  • No products in the cart.