Upskilling is becoming more essential for creatives in an uncertain landscape. Everyone from graphic designers to 3D animators can learn extra skills to give them multidisciplinary experience and make them more versatile. Coursera Plus is one of the online art course providers we recommend in our guide. It has thousands of creative courses – and it is currently discounted by a whopping 50%.
It works pretty simply, you pay annually (or monthly) to access the range of over 10,000 courses, meaning you can complete as many as you like – and the annual sub is currently just $199, reduced from the usual $399. But that offer ends today (June 30) – from tomorrow it moves to a 40% discount for the month of July. See details below:
Fox News finished primetime ahead of two broadcast networks, NBC (2.5 million) and CBS (2.1 million), in total viewers. There is a slow attrition of viewers across the cable news space as the Middle East conflict stabilizes, but Anderson Cooper‘s presence in the region and broadcasting outside of his 8 p.m. ET slot has helped lift some of the network’s shows, such as The Arena with Kasie Hunt. CNN’s 4 p.m. ET show, which he co-anchored. It was the network’s second-most-watched show of the day in total viewers, falling only behind the 10 a.m. ET hour of The Situation Room with Wolf Blitzer, which had 673,000 total viewers.
25-54 Demographic (Live+SD x 1,000)
Total Day: FNC: 273 | CNN: 92 | MSNBC: 73
Prime: FNC: 381 | CNN: 134 | MSNBC: 117
FNC: CNN: MSNBC: 4PM Cain:
314 Hunt/Cooper:
122 Wallace:
95 5PM Five:
440 Tapper:
113 Wallace:
69 6PM Baier:
353 Tapper:
110 Melber:
77 7PM Ingraham:
330 Burnett:
82 Weeknight:
96 8PM Watters:
410 Cooper:
101 Hayes:
81 9PM Hannity:
337 Collins:
136 Psaki:
132 10PM Gutfeld!:
396 Phillip:
167 O’Donnell:
137 11PM Gallgher:
271 Coates:
83 Ruhle:
72
Total Viewers (Live+SD x 1,000)
Total Day: FNC: 2.160 | CNN: 481 | MSNBC: 741
Prime: FNC: 3.223 | CNN: 622 | MSNBC: 1.165
FNC: CNN: MSNBC: 4PM Cain:
2.380 Hunt/Cooper:
659 Wallace:
1.104 5PM Five:
4.069 Tapper:
630 Wallace:
1.109 6PM Baier:
3.072 Tapper:
559 Melber:
918 7PM Ingraham:
2.893 Burnett:
623 Weeknight:
838 8PM Watters:
3.548 Cooper:
632 Hayes:
1.009 9PM Hannity:
3.006 Collins:
633 Psaki:
1.125 10PM Gutfeld!:
3.115 Phillip:
601 O’Donnell:
1.361 11PM Gallagher:
1.734 Coates:
380 Ruhle:
736
Source link
Everything considered goofy and broken about AI images fits right in with Dream Recorder, because dreams are kind of goofy and broken too. As we know, dreams are difficult to explain, but you would probably know one when you see it – illogical physics, environments that blend indiscriminately into each other, bizarre anti-narratives. “AI generates images by filling in blanks, stitching together fragments of learned data into something coherent but often strange. That process mirrors how our minds construct dreams: nonlinear, symbolic, filled with distortions and substitutions,” says Modem.
The actual look of the Dream Recorder is a mix between something clunkily analogue and utopian in its ergonomics, quite like a piece of otherworldly technology from a David Cronenberg movie (particularly the smooth, cartilage-inspired design of the UmbiCords in his 1999 movie eXistenZ). “We wanted to avoid the cold, clinical aesthetic that usually defines tech objects,” says Modem. Inspired by the discourse around “ambient computing” (environments where technology is seamlessly integrated into our surroundings, becoming an invisible part of our lives, such as an Alexa or GPS), the design of the Dream Recorder is chameleonic, an inoffensive part of the furniture – in other words, it exists in the background just like our unconscious thinking. “3D printing gave us the freedom to experiment with these nuances in shape and texture, allowing us to move away from rigid symmetry and toward something more tactile, more grown than manufactured,” says Modem.
Modem describes the bedroom as a “phone-free sanctuary” and its Dream Recorder is an extension of that sacredness. Aiming to protect the mental and physical boundary between rest and distraction, the device has an comfortable presence – a gentle glow in the dark. When training the generative AI’s visual identity, Modem collaborated with artist Alexis Jamet, whose distinctive style features blurry gradients, lo-fi textures, and atmospheres that feel like memories. “We weren’t interested in hyperrealism or slick digital sheen. We wanted something that felt tactile, fallible, almost emotional,” says Modem. “To achieve that, we developed a lightweight on-device post-processing method that deliberately degrades the AI-generated output – after all, we don’t dream in HD.” Maybe the next time you wake up in a cold sweat or with a smile on your face, you won’t need to attempt to explain what you just experienced to your loved one or coworkers – the Dream Recorder will be waiting to make your dreams come true.
We’ve updated our page on Agentforce with details on Salesforce’s Agentforce 3 announcement.
Since its launch at the Dreamforce conference in September 2024, Salesforce Agentforce changed the conversation around AI, customer experience and customer service. Shortly after the launch, Salesforce announced it was hiring more than 1,000 employees to meet the demand for Agentforce.
At the time of the Agentforce announcement, Salesforce CEO Marc Benioff called Agentforce “the third wave of AI — advancing beyond copilots to a new era of highly accurate, low-hallucination intelligent agents that actively drive customer success.”
The platform is so important to Salesforce that Benioff declared at the Agentforce launch, “The only thing we’re going to do at Salesforce is Agentforce.”
Listen to an AI-generated audio Deep Dive of Agentforce based on the contents of this article.
Ask MarTechBot your questions about Salesforce Agentforce and agentic AI. Ask it how companies are using Agentforce, for Agentforce success stories and more.
MarTechBot is thinking …
Agentforce is essentially a platform layer in the Salesforce ecosystem. Agentforce bots are generative AI bots, but unlike popular genAI platforms like ChatGPT and Google Gemini, Salesforce says its Agentforce bots can be trusted to take action on their own.
Unlike earlier chatbots, AI agents do not follow a prescribed pattern of answers to common queries, but are capable of responding more flexibly to questions or issues because they are trained on internal data and are capable of drawing on the support of a large language model (LLM). Salesforce’s agentic AI, for example, can be trained on data in Salesforce Data Cloud.
Agentforce combines three major Salesforce tools — Agent Builder, Model Builder and Prompt Builder — to deliver out-of-the-box bots that work across industries. The bots are customized using inexpensive, low-code tools.
The initial agents created with Agentforce include:
General availability for Agentforce One began on Oct. 25, 2024, with pricing starting at $2 per conversation, with discounts available to large users
Salesforce also announced the Agentforce Partner Network at the launch. It lets other organizations deploy third-party agents or use third-party actions when designing custom agents in Agentforce.
Dig deeper: Customer experience management in the age of agentic AI
During the initial Agentforce announcement at Dreamforce, Benioff said Agentforce is designed for customers who don’t know much about AI or building bots.
“They are not going to have to be an expert in an LLM, they are not going to have to understand all of these deep capabilities that you would need to know to be a computer scientist,” he said. This is “AI for the rest of us.”
Agentforce is powered by an upgraded version of Salesforce’s Einstein AI called Atlas Reasoning Engine, which is “designed to simulate how humans think and plan.”
At the announcement, Salesforce drew a comparison to self-driving cars. Agentforce can interpret data to adapt to conditions in real time and can act independently within a company’s guardrails.
Benioff said Agentforce has the lowest hallucination rate of any generative AI but did not say what that rate is.
Data Cloud provides the customer data that grounds Agentforce, making agents more contextually aware, knowledgeable and adaptable to customer needs. Data Cloud connects, unifies and harmonizes customer data and metadata, giving agents access to the exact information they need to deliver precise, contextual responses. It makes structured and unstructured data – including emails, transcripts and PDFs – easily digestible by large language models (LLMs). The RAG functionality allows users to upload files and URLs as data sources for their agent.
Agentforce is deeply integrated with Salesforce Customer 360, leveraging applications like sales, service, marketing and commerce. A complete view of the customer is critical to seamless hand-offs to human agents with the full conversation history.
As part of its Agentforce 3 announcement, Salesforce said Heroku Managed Inference and AppLink will make it fast and easy to deploy, register, maintain and connect custom MCP servers. With Heroku’s secure infrastructure and DevOps automation, developers can bring trusted custom actions to Agentforce.
MuleSoft is an integration, automation and API management platform. It extends Agentforce capabilities by enabling Salesforce developers and admins to leverage their APIs and bring in third-party data.
As part of the Agentforce 3 announcement and its support for model context protocol (MCP), Salesforce introduced new MCP connectors for MuleSoft that will convert any API and integration into an agent-ready asset, complete with security policies, activity tracing and traffic controls.
Slack is a central point for employees to interact with Agentforce. They can access information, trigger workflows and manage tasks directly within Slack, which means they don’t have to jump between applications.
Agentforce also uses Slack Enterprise Search to learn from conversations in Slack channels and direct messages. This helps agents provide more relevant responses and actions based on the context of conversations.
Tableau Einstein is an AI-powered visual analytics platform that includes Agentforce. It is designed to intelligently embed data and insights across the flow of work and encourages employees to use data to drive action.
Dig deeper: 4 key features in Salesforce’s Agentforce 2.0
In December 2024, Salesforce announced The Testing Center, which lets businesses test their AI agents before deployment. The Testing Center gives Agentforce users an opportunity to ensure the agents follow instructions, stay factual and work quickly. This helps mitigate risks and ensures agents perform consistently and reliably. It also provides tools that make it easier to test, audit and scale agents.
Also announced in December 2024, this library spans key business applications like CRM, Slack and Tableau, and integrations from AppExchange partners. These ready-to-use skills simplify the process of building tailored AI agents. The library has pre-built Slack Actions like Create Canvas and Message Channel, which let teams create agents for their Slack workflows.
Here are some examples:
In January 2025, Salesforce used the National Retail Federation (NRF) conference to announce Agentforce for Retail, which includes a library of pre-built agent skills relevant to retail. It is aimed at facilitating the creation of AI agents by retail brands to supply what Salesforce is calling “digital labor.”
The new skills available in Agentforce for Retail include:
Salesforce also announced Retail Cloud with Modern POS, a new offering that brings online and offline inventory data together on a single platform.
Dig deeper: MarTech launches Agentforce for Retail
Salesforce itself is using Agentforce internally. It says agents at help.salesforce.com now handle 83% of customer support queries independently, with human escalations dropping by 50% in the two weeks after it was implemented.
RBC uses Atlas and Data Cloud with Agentforce to help its financial advisors. Salesforce Data Cloud integrates all RBC client data into a unified system, giving advisors a comprehensive view of their clients. Atlas can answer complex client questions using multistep reasoning. It also pulls from RBC’s data and business logic to deliver a reliable answer.
Accenture uses Agentforce to give employees quick updates on important accounts. It helps automate document retrieval and makes recommendations for solving business problems. It also streamlines proposal creation and visualization and enhances employee collaboration by providing real-time updates, retrieving documents, offering proactive recommendations and creating proposals.
Other companies publicly identified by Salesforce as Agentfore customers include:
Dig deeper: SharkNinja embarks on its Salesforce AI journey
Salesforce customers will be able to build Agentforce agents using Googlr Gemini and deploy Salesforce on Google Cloud, thanks to an expanded partnership between the two compnies.
One example of the expanded partnership is an integration between Grounding with Google Search through Vertex AI and the zero copy partnership between Salesforce Data Cloud and Google BigQuery. This integration will give Agentforce agents the ability to reference up-to-the-minute data, news, current events and credible citations, substantially enhancing their contextual awareness and ability to deliver accurate, evidence-backed responses.
In a real-worlkd example from supply chain management and logistics, an agent built with Agentforce could track shipments and monitor inventory levels in Salesforce Commerce Cloud, and proactively identify potential disruptions using real-time data from Google Search, including weather conditions, port congestion and geopolitical events.
Before the end of 2025, Google’s Gemini models will be available for prompt building and reasoning directly within Agentforce. This will provide users with agents with multi-modal capabilities, expanded contextual understanding and reasoning and increased speed and efficiency.
Additionally, businesses will be able to use Salesforce’s unified platform (Agentforce, Data Cloud and Customer 360) on Google Cloud’s highly secure, AI-optimized infrastructure, benefiting from features like dynamic grounding, zero data retention and toxicity detection provided by Salesforce’s Einstein Trust Layer.
Once Salesforce products are available on Google Cloud, customers will also have the ability to procure Salesforce offerings through the Google Cloud Marketplace, the companies announced.
Dig deeper: Salesforce & Microsoft square off with new AI sales agents
Salesforce announced in April 2025 it is launching Agentforce for Field Service, its digital labor platform built to augment dispatchers and technicians by eliminating scheduling bottlenecks and tackling routine, time-consuming tasks.
Accord to research conducted by Salesforce, inefficient scheduling is the No. 1 drain on field service teams. It’s estimated skilled tradespeople and technicians waste nearly an entire workday each week on administrative work — time that should go toward hands-on repairs and customer service.
Agentforce integrates into existing data systems and user interfaces, autonomously scheduling appointments, assisting with filling schedule gaps, troubleshooting in real time and summarizing job reports. It also features audio playback and natural language voice commands, so field technicians can conveniently interact with Agentforce and consume information while on the move — boosting productivity, safety and response times.
In early May 2025, Sales announced Agentforce for HR Service, a set of AI-powered capabilities embedded directly into Salesforce’s HR Service platform.
Agentforce for HR Service includes a suite of out-of-the-box topics – broad categories of tasks AI agents perform — and actions-specific steps or tools the AI agent uses to complete those tasks.
Agentforce for HR Service lets employees perform common HR support tasks conversationally — right where they spend most of their time, inside applkcations like Slack or their employee portal — without needing to file tickets, dig through policy documents or move between multiple systems.
The AI agents are grounded in company data, knowledge articles and policies, and integrate with leading Human Resource Information Systems (HRIS) and Human Capital Management (HCM).
Among the capabilities of Agentforce for HR Service:
In mid-May 2025, Salesforce unveiled a set of governance, security and compliance capabilities to help organizations deploy trusted AI agents.
These capabilities allow customers to govern data across the entire Salesforce platform, whether from inside or outside of Salesforce applications.
Salesforce uses several products to create trust and data governance, including Agentforce, Salesforce Data Cloud, MuleSoft and Trusted Services (including capabilities from the recently acquired Own, Shield, and the Trust Layer).
Salesforce says the capabilities deliver a fully integrated, enterprise-grade foundation for trusted, governed and secure AI, in contrast to fragmented point solutions available elsewhere. Agentforce runs natively on the platform, providing control, visibility and assurance at every stage of agent deployment.
The capabilities range from Zero Copy data access and built-in policy enforcement to real-time monitoring and AI behavior transparency. They include:
There are also capabilities to help developers build and test Agentforce agents before they go into production.
Read more about Salesforce’s trust and governance initiatives for Agentforce.
Salesforce intially launched Agentforce with a conversational pricing model of $2 per conversation. In May 2025, Salesforce introduced three new pricing models:
Flex Credits offer customers flexibility with a consumption-based model that Salesforce says aligns cost with business outcomes. Businesses only pay for the exact actions Agentforce performs, which could include updating customer records, automating complex workflows or resolving cases. Each of these actions consumes 20 Flex Credits ($0.10 per action). Flex Credits are available in packs of 100,000 credits ($500).
The new Flex Agreement allows organizations to manage both human and digital labor and shift their investments between user licenses or digital labor according to their business priorities. The Flex Agreement converts user licenses into Flex Credits, or Flex Credits into new user licenses, for exploring new, value-generating use cases.
New Agentforce user licenses and add-ons help bring Agentforce capabilities to every employee, offering unlimited employee-facing agent usage in a per-user, per-month (PUPM) pricing model.
Get more details on the Agentforce pricing models.
In May 2025, Salesforce introduced Agentforce for Financial Services, a set of pre-built, role-based AI agent templates for financial services teams. The agents can handfe key front-office tasks like preparing for an investment review, replacing a lost credit card, surfacing relevant loan options and other administrative burdens that detract from more valuable customer engagement.
Agentforce is grounded in a firm’s data, workflows and compliance controls so every action conforms with internal policies and regulatory requirements. Agentforce is natively embedded in Salesforce’s Financial Services Cloud, allowing human and digital workers to operate from the same deeply unified platform in a way that feels intuitive and trustworthy.
Agentforce for Financial Services includes pre-built templates for:
Each pre-built template includes Topics, which guide agent behavior, and Actions that give agents the ability to take action specific to financial services jobs. Firms can customize and expand these agents with Agentforce to reflect their specific processes, guardrails, policies and service models using a declarative, no-code environment.
In June 2025, Salesforce introduced Marketing Cloud Next, which embeds autonomous AI agents across the entire customer funnel.
Salesforce said the product represents a shift from traditional campaign-based marketing to agentic marketing, where AI agents act independently to execute campaigns, personalize customer interactions and optimize performance.
These AI agents can interpret business goals and carry out tasks across marketing, sales, service and commerce functions. For example, marketers can input a goal — such as raising awareness of a new product among top customers — and the AI will build and launch the campaign, manage customer journeys and personalize outreach across channels like email, SMS and web.
Among the features on Marketing Cloud Next are Agentforce Campaign Creation and Agentforce Web Curation.
In late June 2025, Salesforce announced Agentforce 3, which focuses on observability and control, in an attempt to solve a problem vexing companies deploying agentic AI: they can’t see what their AI agents are doing.
Central to Agentforce 3 is Agentforce Command Center. Command Center, which is built into Agentforce Studio, is an observability solution that delivers a single pane of glass to monitor agent health, measure performance and optimize outcomes.
Teams can use Command Center to analyze every AI agent interaction, drill into specific moments, understand trends in usage and see AI-powered recommendations for tagged conversation types, all of which can be used to continuously improve the agent experience.
As part of the Agentforce 3 announcement, Salesforce said Agentforce wouldsoon include a native model context protocol (MCP) client, allowing Agentforce agents to connect to any MCP-compliant server without any custom code. This enables access to enterprise tools, prompts and resources — governed by existing security policies.
Agentforce customers will be able to discover MCP servers from more than 30 partners through AgentExchange. Launch MCP partners include AWS, Box, Cisco, Google Cloud, IBM, Notion, PayPal, Stripe, Teradata and WRITER.
Additional Agentforce 3 enhancements include improvements to the Agentforce architecture to reduce latency, additional support for geographies and additional language support.
Read more details on the Agentforce 3 announcement.
MarTech is owned by Semrush. We remain committed to providing high-quality coverage of marketing topics. Unless otherwise noted, this page’s content was written by either an employee or a paid contractor of Semrush Inc.
If you’re into 3D printing, then you’ll know that using an AMS (automated material system) for multicolour printing is extremely time-consuming. While you can get amazing results using some of the best 3D printers, I’ve found that sometimes, printing smaller features such as claws and eyes in a specific colour isn’t always worth the extra 5+ hours of print time – so I use acrylic markers instead.
The set I currently use is by NICETY, and it’s on sale right now with 27% off, down to just £16 over at Amazon. For this price, you get a total of 58 acrylic paint pens covering pretty much every colour you could ever need. I recently used these pens in my tutorial on how to 3D print cosplay props.
Image 1 of 3
(Image credit: Future)(Image credit: Future)(Image credit: Future)
Today’s best deals on 3D printers
Google AI Mode, which officially launched in May 2025 and is now available to all U.S. users without a waitlist, represents a significant step forward in how we engage with search.
Powered by Gemini 2.5, this new interface moves beyond AI Overviews by introducing a persistent, conversational assistant that blends AI-generated insights with traditional search results.
Users can toggle between classic results and AI-driven summaries, follow up on queries, and explore longer, more exploratory conversations, all within a single interface.
Unlike AI Overviews or the earlier Search Generative Experience (SGE), which provided a single AI-generated answer for a traditional search query, AI Mode is more similar to ChatGPT in that it fosters a conversational approach to finding answers.
This marks a change in how people interact with search, moving from short, isolated keywords to more natural prompts that sound like how we talk and think.
AI Mode supports rich interactions and longer queries, encouraging a deeper and more nuanced engagement with information. And when user behavior shifts, advertisers must adapt how they reach users with relevant solutions and offers.
Think back to how Enhanced Campaigns forced advertisers to get ready for the explosion in mobile device usage.
We’re now at another junction where advertisers and Google must work together to evolve how we operate to remain successful. That means reconsidering everything from targeting and attribution to monetization and ad design.
AI Mode Interface (Screenshot from Google, June 2025)
In this post, I share my thoughts on what AI Mode signals for the future of search, how it challenges long-standing digital advertising models, and why marketers need to adapt fast or risk being left behind.
Is Google pushing AI Mode because it sees an opportunity or because it’s responding to pressure from OpenAI and others? The answer is likely both.
Google’s technical leadership is well-established.
DeepMind, a Google company, helped invent the transformer model that underpins GPT. Its Gemini family of models has matured rapidly.
At Google Marketing Live 2025, Sundar Pichai stated that Gemini had taken the lead as the top-performing model, a claim supported by LM Arena’s leaderboard.
Still, Google moves cautiously. As a market leader under regulatory scrutiny, it can’t afford missteps.
The innovation is real, but so is the strategy to protect its dominance by making AI part of its core products before others can take the lead.
I believe Google’s technology is among the best in the world. However, as the company is in the spotlight, they have to be more measured.
Regulatory scrutiny, scale, and legacy expectations mean it can’t move as fast as emerging players, but that doesn’t mean it will always be chasing the lead.
How users like to find answers is changing from clicking around on a search results page to interacting with an AI assistant.
This evolution from search engine to answer engine introduces a new layer of complexity for advertisers. Prompts in AI Mode aren’t just text; they’re conversations rich with personal context and memory.
Take a user engaging in a long session with AI Mode. Their conversation might include several prompts in a row like this:
The assistant understands the goal and tailors responses to match medical considerations, intent, and emotional tone.
It might include surface stability shoes, recommended inserts, and even factor in training timelines or the expected weather in the city where the marathon will take place.
Now contrast that with a short prompt: “running shoes.”
Simple on the surface, except the assistant remembers that just yesterday, I was at the Adidas store talking to a clerk about shoe fit via my bee.computer wearable, and I used my Ray-Ban Meta glasses to snap a few images of colors I liked.
While this use case is not quite there yet in the real world, I am personally using this technology now, and it’s just a matter of time until all the pieces are connected and the advertiser scenario I described will become real.
Then we’ll see the assistant pick up right where I left off, using multimodal memory to enrich the response with past conversations and visual preferences.
Neither of these interactions can be matched with traditional keyword-based targeting. The assistant’s memory and personalization turn every query into a unique moment.
For advertisers, it’s not just about what was typed; it’s about what the assistant knows.
This creates a richer opportunity for advertisers, but there is a challenge related to targeting because Google Ads was built for keyword advertising, not prompt advertising – and this creates a disconnect.
→ Curious how marketers are adapting to these shifts? Check out the 2025 State Of AI In Marketing report for the data and insights.
Google Ads was initially built around a simple idea: Match ads to user searches through keywords.
Advertisers bid on terms users might type into the search bar (like “running shoes” or “cheap flights”), and the system will serve relevant ads based on those inputs.
But AI Mode is changing the language of search. Instead of short, isolated keywords, users are starting to use full, conversational prompts that reflect how they naturally speak.
These prompts are often longer, more specific, and packed with nuance that the original ad system wasn’t designed to handle.
To keep things running, Google has introduced a behind-the-scenes workaround: “synthetic keywords.”
These are machine-generated representations that attempt to map modern prompts back into the keyword framework advertisers still rely on. It’s a clever patch, but ultimately a temporary one.
As prompts continue to evolve in complexity and variety, and as memory and personalization shape every query, the keyword as a stable targeting anchor is becoming harder to rely on.
That puts pressure on the entire ad ecosystem. The old model is still functioning, but it’s increasingly out of sync with how people search.
A new system, one built natively for prompts, context, and memory, will eventually need to take its place.
The shift toward AI-assisted browsing brings another major challenge: fewer clicks.
If users get what they need from the assistant itself, the need to visit websites diminishes, weakening the foundations of the cost-per-click (CPC) business model.
Slide by Microsoft at Accelerate Roadshow LA, June 2025
But clicks will be more relevant because, unlike in the past, where a click was a user’s initial exploration of your offer, they will now be better informed and further along in their research by the time they visit your site for the first time.
Microsoft research found that purchasing behaviors increased by 53% within 30 minutes of a Copilot interaction, underscoring just how powerful, timely, and AI-embedded suggestions can be.
To stay relevant, ads must feel like part of the conversation. They can’t be disruptive or detached. They need to be embedded, responsive, and helpful, appearing when and where they make the most sense.
Newer performance data shows that ad engagement doubled in some formats when served through Copilot, especially in PMax-powered Shopping and Multimedia Ads.
Crucially, Microsoft has dialed back the volume of ad impressions in Copilot, choosing instead to show ads only when they’re predicted to be highly relevant and useful.
The result? Fewer, better-placed ads that drive stronger outcomes, a model that hints at where Google AI Mode could be headed.
Google has done this before. Its introduction of AdWords transformed ads from flashy banners into useful information. AI Mode demands a similar evolution, one that turns helpfulness into performance.
So, if the traditional way Google makes money becomes broken, let’s look at some options for how they might bridge the gap.
→ Understand where marketing teams are realizing value and where key gaps remain. Check out the 2025 State Of AI In Marketing whitepaper.
The most frustrating part for consumers using AI agents to find something to buy is the final step after determining what they want.
Now, they need to hunt for where to buy it, enter a credit card, and deal with the usual minutiae of buying something online.
A better user experience, especially for smaller purchases, would be to tell the agent, “I like it, buy it!” and have the item arrive at your doorstep the next day.
While this zero-click scenario is the best user experience, it is also the most problematic in a CPC world.
This opens the door for reconsidering affiliate and commission-based advertising models. Instead of paying for attention, advertisers pay for action.
Ads become decision-making partners, not just traffic generators. It’s a better fit for how assistants work: focused, efficient, and user-first.
While this wouldn’t be Google’s first attempt at commission-based monetization (previous efforts, such as Buy on Google, Shopping Actions, and Google Express, ultimately shut down due to limited merchant adoption and weak consumer uptake), those models lacked the personalized context that AI Mode now enables.
Even vertical-specific experiments like commission bidding for Hotel Price Ads (retired in 2024) followed the same pattern: strong in theory, but missing the behavioral depth to sustain engagement.
With memory-driven prompts, real-time user needs, and multimodal signals in play, the conditions may finally be right for performance-based pricing to scale in a meaningful, consumer-aligned way.
Monetizing AI-powered search is a hot topic. Startups like Neeva by Sridhar Ramaswamy (Former Google Ads Chief) attempted to replace ads with subscriptions, but user adoption fell short.
Even OpenAI, with its paid ChatGPT Pro tier, sees a vast majority of users opting for free access.
The pattern is clear: Most users won’t pay for general-purpose search tools. Even companies leading in AI anticipate that advertising will remain the dominant revenue stream.
Google’s ad model, tested and refined for decades, is still the best-positioned approach – if it can evolve to match the new user behavior.
Google has already said it will have ads in AI mode.
To maximize the likelihood of your ads appearing in this environment, it’s advisable to utilize Google’s AI-centric tools, including AI Max in search campaigns, Performance Max, and Demand Gen.
Employing broad match keywords is also crucial, as they facilitate connections with conversational prompts rather than traditional keywords.
However, with the potential decrease in click-through rates, a pertinent question arises: Can fewer clicks on ads sustain the revenue model?
Despite this challenge, I anticipate that advertising will remain the primary revenue stream, even within AI Mode.
It’s noteworthy that OpenAI’s CEO, Sam Altman, has expressed reservations about incorporating ads into AI experiences.
In a conversation with Ben Thompson, Altman stated:
“Currently, I am more excited to figure out how we can charge people a lot of money for a really great automated software engineer or other kind of agent than I am making some number of dimes with an advertising-based model… I kinda just don’t like ads that much.”
Similarly, Google’s co-founders, Larry Page and Sergey Brin, initially opposed the idea of advertising on their search engine. In their 1998 research paper, “The Anatomy of a Large-Scale Hypertextual Web Search Engine,” they wrote:
“We expect that advertising-funded search engines will be inherently biased towards the advertisers and away from the needs of the consumers.”
Despite these initial reservations, both OpenAI and Google have recognized the practicalities of monetization. Google makes nearly 78% of its revenue from ads as of 2024, illustrating its evolution from the original stance of its founders.
So, while the methods and philosophies around advertising in AI experiences have evolved, the necessity for effective monetization strategies remains paramount.
Soon, helping consumers at the moment of relevance won’t be about search and keywords anymore; it’ll be about context, and AI-powered interactions driven by memory, intent, and dialogue.
The early signals are promising: Users respond better when ads are useful, not intrusive.
Microsoft’s experience with Copilot shows that when generative systems deliver fewer but more relevant ads, engagement and conversions rise.
Google’s opportunity is to take those lessons further, baking utility and timing into its AI-native monetization engine.
It’s not about building the flashiest assistant; it’s about earning trust at the moments that matter.
If the assistant can deliver value and drive outcomes without breaking the flow, that’s the model that wins.
I have no doubt that Google and other ad platforms will find ways to appropriately monetize these advertising opportunities, even if there will be fewer impressions for each consumer journey.
The fundamentals of advertising at the moment of relevance haven’t changed, but our tactics will need to evolve fast. Prompts, not keywords, are the new starting point – and that changes the game.
More Resources:
Featured Image: Lana Sham/Shutterstock
But it’s not a sure thing.
In melding Omnicom and IPG, the deal will create the world’s largest advertising group. In some countries, the new conglomerate will take on an enormous share of the media-buying and advertising market. In New Zealand, for example, the combined entity could gain a 55% market share overnight, according to industry advocacy group Independent Media Agencies New Zealand.
So before the champagne bottles can be uncorked, the deal has to pass muster with a number of competition regulators in key markets around the globe, including Britain and Australia.
Let’s look at some of the relevant players, and how their moves might yet put a kink in the Omnicom-IPG mega deal.
Both parties have agreed to abide by a “consent order” issued by the FTC which (in theory) will prevent the new company from exerting undue influence over publishers in the U.S. But the highly unusual agreement isn’t the end of the story.
The FTC’s now opened the floor to allow interested parties to formally make public comments on the merger for 30 days; final approval will come once that window closes.
The public comment stage is unlikely to derail the deal by itself. “While the FTC is obligated to invite comments, it isn’t required to listen to them,” said Ray Seilie, litigator at Kinsella Holley Iser Kump Steinsapir LLP (KHIKS), in an email.
But it could expose the company to further political pressure from the current presidential administration and its outriders — who might feel the consent order doesn’t go far enough in enforcing their idea of political neutrality on advertisers.
“It will be interesting to see which groups choose to submit comments and what positions they end up taking. But it would be unusual, especially with this administration, for those comments to result in a confession of error and a revision of the consent order,” said Seilie.
“We continue to look forward to obtaining the remaining regulatory approvals and closing in the second half of this year, consistent with our expectations when we announced this transaction,” Omnicom CEO John Wren said in a statement earlier this week. Top of the list? Britain’s Competitions Markets Authority (CMA). In 2023, the U.K. accounted for 8.5% of Omnicom’s global revenues and 6.3% of IPG’s, according to analysis by Campaign.
Though its involvement in the merger is technically voluntary, the CMA has already kicked off a 40-day formal inquiry into the deal. That can lead to a number of outcomes. The Authority might stage a deeper investigation, which could take as long as 18 months. It might prohibit the merger outright in Britain, or it might accept it — with or without strings attached.
Paolo Palmigiano, head of the competition, trade and foreign investment practice at law firm Taylor Wessing, said that should the CMA choose to add conditions to its approval, they’re more likely to be structural, rather than try to restrict its investment choices like the FTC’s order does.
“It normally is a remedy like divestment from a unit or something similar,” he said.
The Australian Competition & Consumer Commission (ACCC) is also looking at the merger, as is the EU Commission. The former’s due to release its findings July 27th.
Meanwhile, 10 competition authorities have already given the nod to Omincom and IPG, including the Competition Commission of India (CCI), as well as similar bodies in New Zealand, China, Japan, Brazil and Singapore.
The FTC, CMA and counterpart organisations do cooperate for cross-border mergers. “It suits [all] parties for the competition authorities to talk, because then it’s faster,” said Palmigiano. Mostly that takes place on an informal basis though in some cases, a regulator might ask for a formal waiver from the companies in question to collaborate with another regulator.
But the bottom line is that each regulator only has jurisdiction over their own market — and approval in one territory doesn’t necessarily mean the dominos will fall Omnicom’s way in another. “Their legal standards are different, so just because the U.S. believes a merger is competitive (subject to the consent order) does not mean that these other countries will automatically reach the same conclusion,” said Seilie.
Not just. Bodies like the FTC and the CMA are independent, but that doesn’t mean they’re untouched by broader political philosophies.
Under former chair Lina Khan, the FTC expanded its use of antitrust law into a broader means of curtailing the power of the tech sector. Under her successor Andrew Ferguson, it has continued that practice. And reading between the lines of his statement on the Omnicom merger, it’s apparent the FTC saw a chance to send a strong signal to the advertising sector about where it places clients’ media dollars.
“We are presented… [with] a troubling history of collusion to the detriment of consumers and the free conduct of American political discourse and elections,” he wrote. “In the absence of any intervention, the proposed acquisition is likely to substantially reduce competition and may enhance the vulnerability to coordinated effects.”
Similarly in recent years, the CMA has been willing to pose difficult questions of acquiring companies. Its reservations delayed Microsoft and Activision’s 2023 merger by several months and sparked a minor political row before eventually greenlighting the deal.
But influence from Britain’s Labour government on the CMA means its attitude toward mergers might become more forgiving. In January, chancellor of the exchequer Rachel Reeves suggested the CMA, alongside other regulators like Ofcom, should be working harder to support the U.K. economy.
“Every regulator, no matter what sector, has a part to play by tearing down the regulatory barriers that hold back growth,” she said at the start of the year.
Shortly afterwards the CMA’s chair Marcus Bokkerink was dismissed and replaced by Doug Gurr, a former Amazon executive. “The reason was to give a clear signal to the regulator: ‘This is the line you have to toe.’ It’s all about growth,” said Palmigiano.
The regulator is still independent. And it still has teeth. Just this week, the CMA said it wanted to set more binding rules around how Google operates search in the U.K. and how publishers’ content is used, including AI Overviews.
So, whether a pro-growth agenda leads to a softer touch in Omnicom’s case remains to be seen. “We haven’t seen that yet,” cautioned Palmigiano.
Update: This article was amended to correctly describe the status of New Zealand’s Commerce Commission’s inquiry.
Zoë Ene trained as an industrial designer at the Rhode Island School of Design (RISD), where she recalls being conflicted with what was being taught than what she knew of her Igbo heritage and its craft philosophies. “As a young designer, I was eager to challenge that gap – probably annoying a few lecturers in the process – and became interested in work that explored the opposite,” she says.
Now, she doesn’t just make work that is culturally interwoven but uses said perspective to document theories, methods and tools that aid her practice and others. Her recent work, the IO stool, is shaped like a kola nut, called isi oji in Igbo, and includes perforated motifs that appear in triangular gaps, giving it a structural impression.
Homenkà is the platform that Zoë founded to shape her design language and lead her closer to deep cultural research. It’s a portmanteau of the Igbo word Omenkà (artist) with the word Home, which reflects the commitment to homegrown approaches to designing objects. “My goal is to make the value of Nigerian traditional design visible, usable, and meaningful in everyday life – especially in Nigerian homes and shared spaces. We’re just getting started, and I’m excited for what’s ahead.”
Her forthcoming work Ncho 01 is a prototype mancala board (also known as oware, ayo, bao, or ncholokoto from which the board gets its name) and is centred on continuing Igbo woodcarving making traditions and embodies the duality of function and beauty that defines much of African object design. “At our booth, people taught each other how to play, shared memories of using stones or digging holes in the earth to create boards as kids, and responded with joy at seeing a familiar game reimagined.”
Reclaiming the narratives of their heritage is not just a bus stop for these designers, they are also looking inward, extending their knowledge to those who care deeply about this practice. With that said, it wouldn’t come as a surprise seeing further innovations of this ilk that would shape the future of design.
In boardrooms and Slack threads alike, “demand generation” and “lead generation” are often used interchangeably, sometimes even by marketers themselves.
But for CMOs making six- and seven-figure budget decisions, lumping the two together is a costly mistake.
On the surface, both strategies aim to generate revenue. But the approach, intent, and impact of each are fundamentally different.
Understanding these differences isn’t just marketing semantics. It’s a strategic imperative.
Whether you’re scaling a SaaS company, leading an enterprise rebrand, or trying to make sense of declining pipeline velocity, the way you approach demand and lead gen can either fuel long-term growth or lock you into a hamster wheel of short-term wins.
Let’s unpack what each of these approaches actually looks like, where they work best, and how to decide which path (or combination of them) is right for your team.
Demand generation isn’t just a top-of-funnel tactic. It’s a full-funnel strategy designed to create awareness, spark interest, and ultimately build desire for your solution, oftentimes before the buyer even knows they need it.
It prioritizes visibility, trust, and education over form-fills and gated assets.
So, what isn’t demand generation?
Demand gen isn’t about chasing contact details.
It’s about shaping buying decisions before the buyer ever enters a sales process.
This strategy leans heavily on value-driven content, community building, media exposure, and delivering information that builds brand affinity over time.
Examples of some commonly used demand generation tactics include:
In demand generation, you’re not asking for the sale. You’re creating an environment where the sale becomes inevitable.
Lead generation is all about conversions, and not in the philosophical sense.
It’s measurable, trackable, and often deeply tied to sales-qualified metrics. You offer something (a whitepaper, webinar, trial) in exchange for something (a name, email, job title).
The focus here is less on brand building and more on pipeline development. It’s tactical, efficient, and often short-term.
That doesn’t make it “bad,” but it does mean you’ll need a strong nurturing process and sales alignment to make it effective.
Common lead generation tactics include:
Opposite of demand gen tactics, lead gen tactics are a bit easier to measure. They’re also easier to misuse.
If you’re not aligning on what constitutes a “qualified lead,” you might end up with a pile of marketing qualified leads (MQLs) that sales ignores.
While the two approaches might feel similar in campaign execution, the intent and measurement couldn’t be more different.
Element
Demand Generation
Lead Generation
Primary Goal
Build interest & educate the market
Capture contact info for nurturing & sales
Buyer Stage
Early to mid-funnel
Mid to late-funnel
KPIs
Brand engagement, direct traffic, pipeline contribution
Form fills, cost-per-lead (CPL), MQL to SQL conversion
Channel Mix
Social content, podcast, YouTube, native ads
Paid search, lead forms, email, retargeting
Attribution Window
Long-Term (30+ days)
Short-term (<30 days)
If you’re measuring demand gen with the same key performance indicators (KPIs) as lead gen, you’re setting yourself up for disappointment.
These strategies operate on different timelines and serve different roles in the buyer journey.
Let’s say you’re in the B2B SaaS space, and your board wants more pipeline, fast. So, you crank up spend on paid search and run gated ebook campaigns.
You get thousands of leads … and sales team closes almost none of them.
Why?
Because those leads weren’t ready to buy. They downloaded an asset, not because they were in-market, but because they were curious. That’s not a sales-qualified lead; it’s a reader.
On the flip side, if you only focus on brand and never collect contact info or move people into a nurture stream, your pipeline may dry up altogether.
Misalignment here causes poor return on investment (ROI), frustrated sales teams, and confusion at the executive level.
And CMOs? You’re the one who gets held accountable.
If you’re stuck in the “more leads, less revenue” loop, demand gen might be the missing piece.
Watch for these tell-tale signs:
In these cases, shifting some of your focus (and budget) toward demand gen can help you break the cycle.
It doesn’t mean you stop generating leads. It means you start warming the market, so the leads that come through are higher intent and closer to revenue.
Lead gen isn’t dead. It just needs context.
For mature markets or lower-cost products with short sales cycles, lead gen can still be incredibly efficient.
It’s also useful when:
If your team excels at lead nurturing and you’re using lead gen to support (not substitute) long-term demand creation, it can drive fast, measurable results.
Just don’t treat it as a long-term growth strategy in isolation.
This isn’t a zero-sum game. The smartest CMOs know how to balance both.
Think of demand gen as fueling interest, and lead gen as capturing it. The two should work in tandem.
Start with demand creation: educate, build trust, and generate awareness in the market. Then, as interest builds, use lead gen strategies to convert that attention into a measurable pipeline.
If you’re only doing one, you’re either leaving money on the table or burning through it too fast.
Here’s where many CMOs get tripped up: trying to measure demand generation with lead generation metrics.
Demand generation is more about contribution to the pipeline, not generating immediate conversions.
For demand gen metrics, you’ll want to take a look at:
Meanwhile, lead gen metrics like CPL and MQL-to-SQL rates are better used in a supplementary way, not as the only measure of success.
And let’s be honest: Attribution will never be perfect. As CMOs, don’t expect your marketing teams to attribute each effort with 100% accuracy. You’d be setting them, and yourself, up for failure in the long run.
Buyers today might see a LinkedIn post, hear a podcast, and Google your brand three weeks later. That journey doesn’t show up in a neat linear model.
So, rather than obsessing over pixel-perfect attribution, focus on momentum. Is pipeline velocity improving? Is your CAC going down over time? Are more of the right buyers coming inbound?
Those are the real signals you should be looking for to understand if your demand gen and lead gen efforts are working.
This isn’t about choosing sides on which strategy to focus on. It’s about choosing alignment on how the two will operate together.
If you’re stuck on which to prioritize, ask yourself the following questions:
Start there. Then, audit your current marketing mix.
You might find that 80% of your spend is on lead generation efforts, but 80% of your growth comes from demand generation channels.
Chasing short-term tactics only squeezes out who’s currently in your marketing funnel.
You need to build a system that creates both interest and intent.
The most effective marketing strategies don’t live behind a gate. They live in conversations, videos, buyer communities, and the minds of decision-makers before they ever hit your website.
That’s what demand gen does best: It plants the seed between prospective customers and your brand.
Lead gen has its role, but without demand gen, it’s like harvesting from a field you never watered.
For today’s CMOs, the real challenge isn’t picking one over the other. It’s learning how to weave them together into a strategy that works for your audience, your sales team, and your business goals.
Because real growth rarely starts with a form fill, but it can end with one.
More Resources:
Featured Image: ra2 studio/Shutterstock
Anchor Kris Laudien is out at Detroit CBS owned station WWJ. A CBS spokesperson told TVSpy, “Kris Laudien is no longer with CBS Detroit. We wish him the very best.” […]
Source link
