Week In Review: The parent company of Pizza Hut and KFC has launched customer-focused AI tools, Omnicom was named OpenAI’s marketing agency, and NewsCorp warns about AI threat against publishers amid loose US regulations.
Brands and Agencies
Yum Brands Launches AI Tools
Yum Brands – the parent company of KFC, Pizza Hut, and Taco Bell – announced it is launching a set of AI tools for customer and employee experiences. The system of AI technologies, named ‘Byte’ will make it easier for customers to customise online orders and aggregate restaurant reviews to inform on-site operations – this includes the use of an AI-powered voice tool. There is also a drive-thru AI voice that takes orders and relays a simplified version back to workers for ease – this tech is reportedly reducing employee turnover.
Omnicom Named OpenAI’s Media Agency
OpenAI has named Omnicom’s PHD Media as its global media agency of record – meaning the company will manage and execute all, or most, of its marketing and advertising campaigns. This partnership is the first of its kind for the AI giant, and will be led by its first-ever chief marketing officer Kate Roach. Under Roach’s leadership, OpenAI has gone all-in with its marketing efforts. Most notably, the company launched a 60-second animated ad featured at this year’s Super Bowl – made in collaboration with Accenture Song.
WPP Profits Drop by Over 70 Percent
WPP’s pre-tax profits dropped 71 percent in the first half of 2025. The holding giant has also cut its interim dividend to 7.5p a share and will conduct a strategic review led by its new CEO Cindy Rose – who is taking over from Mark Read on September 1. Most major marketing holding companies have reported less than fruitful economic growth. WPP’s struggles have been widely attributed to the widespread rise of AI. In response, the company increased its annual AI investment to £300m.
Publicis Groupe NZ Introduces Data Intelligence Hub
Publicis Groupe’s branch in New Zealand has launched a new Data Intelligence Hub (DIH). The tool enhances the agency’s market research, user experience, and web analytics. Built into the tools is support for data unification, cloud consultation, building data cleanrooms, the development of predictive models, econometric modelling and Gen AI applications. Emily Isle, data intelligence hub Lead, said: “The Hub has been designed to meet the growing demand for an integrated, data-led service model that can deliver fit-for-purpose data and technology solutions tailored to each client’s unique business or organisational needs.”
Eversana Unveils AI Marketing Agency
Eversana, a global commercial services provider for the life sciences industry, is launching an AI marketing agency. The AI Agency will be built on-top of Google Cloud technology and is 80 percent automated, utilising agentic AI to carry out tasks without human input. The services firm cuts costs and speeds up operations for clients. “This is a remarkable moment for our industry. To be very clear, this is not comparable to just using AI tools to enhance copywriting or content creation like so many agencies already do. With Google Cloud, we are deploying a full-scale AI-powered agency platform,” said Faruk Capan, EVERSANA’s chief information officer.
Omnicom and IPG Merger Finalises
The merger between Omnicom and IPG has been passed by the UK regulatory arm, bringing the intertwining of both holding companies closer. The Competition and Markets Authority said the deal would not have to go to a more in-depth probe, meaning it will likely close within coming months. The deal has widely been seen as a strategic move against the backdrop of AI disrupting the advertising industry. IPG’s CEO Philippe Krakowsky said the merger would enable the company to “meet the evolving needs of clients in a consumer and media landscape being transformed by technology and data.”
Media
NewsCorp Warns About AI Theft
In a statement following its full-year financial results, the news conglomerate owned by Rupert Murdoch, NewsCorp, discussed the growing threat of AI to publishing. NewsCorp’s CEO Robert Thomson has retaliated against President Donald Trump’s calls for looser AI regulations in the US. Many publishers, however, have spoken out against AI companies for using content to train AI models without permission. The Wall Street Journal, one of NewsCorp’s publications, is currently facing a lawsuit from the President himself. “The AI age must cherish the value of intellectual property if we are collectively to realise our potential.” Robert Thomson, chief executive, NewsCorp, said. The executive continued that even US President Donald Trump’s works are at risk of being stolen by AI systems. Thomson shared that the President’s books “are being consumed by AI engines which profit from his thoughts by cannibalising his concepts.”
ElevenLabs Creates AI Music Generator
ElevenLabs, the text-to-speech AI platform, released a music generating model. It includes features such as adjusting the style, ranging from techno to pop to synthwave. In a chat box, users can type adjustments to the song such as “mention ElevenLabs in the intro.” The tool then generates a song complete with multiple verses. ElevenLabs shared example songs including one offering “dreamy slow indie rock with reverb vocals, phased guitars, and nostalgic vibe” and another with the descriptor “slow latin reggaeton with charismatic Spanish MC, dancehall vibes.”The model has full commercial rights, the startup said. ElevenLabs established deals with digital platforms for musicians Merlin Network and Kobalt Music Group.
Bloomsbury Publishing Considers AI Deals
Bloomsbury Publishing is looking into AI licensing deals – allowing AI developers to use the publisher’s books to train AI models. Authors with works published by Bloomsbury have the option for their books to be included, which would include receiving one-fifth of the deal’s value, The Bookseller reported. By establishing a licensing deal, Bloomsbury would join many publishers who have done the same. Last year, book publisher HarperCollins signed a deal with Microsoft which gave non-fiction authors the option to receive a sum in exchange for their work being used in AI model training. Publishers like NewsCorp, the Financial Times, and the Associated Press also hold deals with AI companies such as OpenAI and Prorata.ai.
Truth Social Tests AI Search Engine
Truth Social, the social media platform owned by US President Donald Trump, began testing an AI search feature called “Truth Search AI.” The feature is powered by AI search engine startup Perplexity. Truth Search AI’s purpose is to provide more context to posts on the platform with citations, similar to the community notes features seen on platforms like X and Facebook. The tool is “intended to enhance the Truth Social platform and exponentially increase the amount of information available to its users,” Trump Media and Technology Group Corp explained in a press release. The feature, however, has already been criticised for holding bias. The sources cited by Truth Search AI are controlled by Truth Social, which turns up Fox News as the sole source for many responses, TechCrunch explained.
Pinterest Forsees More Growth From AI Advertising Capabilities
Pinterest released its second quarter 2025 results (the quarter ending on June 30, 2025), reporting a 17 percent boost in revenue. The company attributed this growth to its expansion of AI technology for advertisers, including its campaign automation platform Performance+. “Three years into our business transformation, I’ve never been more confident in Pinterest’s ability to deliver for our users and advertisers.” Bill Ready, chief executive officer of Pinterest, said in the company’s financial report. “We’ve found our best product market fit ever by becoming a personalised shopping destination for users and an AI-powered performance platform for advertisers.” Pinterest also saw a surge in Gen Z users, which now make up more than half of its user base.
Cloudflare Accuses Perplexity of Bypassing Anti-Scraping Rules
Internet content provider Cloudflare has accused the AI startup Perplexity of scraping online content despite being blocked from doing so. In September last year, Cloudflare announced it would be automatically blocking AI crawlers and agents unless website owners purposefully opt-in to having their content used. A month ago, this offering developed to include a ‘pay-per-crawl’ feature that would let publishers get paid from their content being taken by AI crawlers and agents. The company claims that Perplexity ignored Robots.txt files that help communicate which part of a website can be accessed by crawlers. The content provider also claims that Perplexity disguised its identity by switching user-agent strings. Perplexity has denied this.
Tech
OpenAI Announces Open Models
OpenAI launched two open-weight AI reasoning models this week. The launch of the two models marks the AI giant’s first open language model released in five years, since GPT-2. OpenAI has predominantly preferred releasing closed models that grant access to businesses through APIs. Many Chinese LLMs – like DeepSeek and Moonshot – are already open models, putting pressure on Western AI companies to offer similar products. OpenAI’s new models are available on AI model platforms such as Hugging Face and Amazon Web Services (AWS).
Mobiquity Launches Agentic Marketing Platform
Mobiquity Technologies Inc has launched CMOne, a fully agentic AI marketing platform. The US-based data intelligence and advertising solutions company said the platform will be unifying organic content, paid media, and conversational engagement – all in one agentic system. Dean Julia, CEO of Mobiquity Technologies, said: “With CMOne, we’re introducing a new operating system for marketing – one that thinks, acts, and grows like your best team member, but works 24/7. It’s not just AI-powered; it’s fully agentic.”
OpenAI to Reach 700m Users
AI giant OpenAI has said it’s set to reach 700m active users on its chatbot platform ChatGPT, a 200m hike since the end of March. The company’s head of product, Nick Turley, wrote in a blog post that the company’s team is “learning, creating, and solving harder problems”. This marks a significant milestone for the parent company of ChatGPT – which launched in November 2022 – sending a shockwave of buzz and AI investment across the globe.
Cohere Announces Agentic Tool for Data Protection
Canada-based AI company Cohere has announced the launch of an agentic AI platform that will keep business data protected from external threats. AI models, especially those used in enterprises, require company data. Now, Cohere’s new platform called North, will utilise AI agents to keep customers’ data safe. The tool has a chat and search feature, summarises meeting transcripts, writes marketing copy, and can access internal and external sources of information. Cohere says North can operate on a business’ infrastructure without needing access to the data itself, unlike other similar tools.
OpenAI Releases ChatGPT-5
OpenAI released its latest model, ChatGPT-5, after months of promotion. This AI system is said to possess stronger reasoning and coding capabilities, with a range of new features. OpenAI shared that GPT-5 provides responses more efficiently than OpenAI o3 and is around 80 percent less likely to include false information in its responses than o3. GPT-5 comes with the ability to change the tone of the AI’s responses, from default to cynic, robot, listener, or nerd. The system can also now be connected to users’ google accounts. With this feature, ChatGPT can help users with tasks like scheduling and managing emails. GPT-5 is becoming available to Plus, Pro, Team, and Free users; and will be available to Enterprise and Edu users next week.
Gartner Releases Top AI Innovations for 2025
Insights platform Gartner released which sectors of AI are seeing the biggest growth. The company outlined these trends in the 2025 Gartner Hype Cycle for Artificial Intelligence, an annual report on the AI landscape. Gartner shared that AI agents and AI-ready data are the fastest-growing sectors in the industry. AI agents, AI systems which can automatically carry out tasks using their own reasoning skills, have become popular assets to enterprises for work tasks. AI-ready data is defined as data sets which are structured to be able to be run through AI models. This helps AI models learn better and better analyse the data at hand. Gartner included that multimodal AI, models trained on different forms of information such as images and videos, will also “become increasingly integral to capability advancement in every application and software product across all industries in the next five years.”
X to Introduce Ads within Platform
Social media platform X will introduce ads within its AI chatbot Grok, the Financial Times reported. The company’s owner Elon Musk announced during a live chat on Wednesday that advertisers will be allowed to pay to appear in the AI chatbot’s query results. “If a user’s trying to solve a problem, then advertising the specific solution would be ideal at that point,” said Musk. Musk also highlighted plans to automate the advertising process and improve targeting for ads. Grok is also planning to introduce a feature that prioritises the best performing, and best looking, ads, and provides a checkout within the chatbot so users can make purchases remaining in the app.
Google Launches Genie 3 for 3D Environments
Google DeepMind, the tech giant’s AI research lab, announced the launch of Genie 3, which can create 3D environments at 720p resolution from text prompts. This level of quality in the new Genie model is an upgrade from Genie 2, which operated at 10 to 20 frames per second instead of Genie 3’s 24 frames per second. Genie 3’s environment generations can stay physically consistent over time because the model remembers what was previously generated. Users will be able to infinitely adapt virtual worlds for industries such as gaming, robotics, training, disaster preparedness, education, and for creating immersive experiences.
Anthropic Bans OpenAI From Using Claude Models
Anthropic has reportedly prohibited OpenAI from accessing its Claude models. Anthropic banned OpenAI after accusing it of violating its terms of service, WIRED reported. OpenAI employees were reportedly using the Claude models to measure the performance of its own models. Using certain application programming interfaces (APIs), OpenAI was allegedly integrating Claude into its own technology. This allowed the company to see how its models compared to Claude in areas such as creative writing and coding. This testing would help the ChatGPT parent company improve its model’s behaviour and how they respond to prompts.
Number of the Week
$155 billion. That’s how much was spent on AI by ‘Big Tech’ already this year. The world’s major tech companies – comprising of Meta, Microsoft, Amazon, and Google’s parent company Alphabet – have all spent the staggering amount on AI since the start of 2025, as reported by the Guardian. The tech companies all reported their annual financial results for the first half of the year, unveiling the not-so surprising fact that they had spent a lot on AI, as desires to outpace one another heats up.
