$META is stepping into the AI search engine arena. The tech giant is reportedly developing its own AI-powered search engine. This move aims to reduce Meta’s reliance on established search platforms like $GOOGL’s Google and $MSFT’s Bing.By building an in-house solution, Meta seeks to control more of its data and user experience. However, this centralization raises eyebrows in the crypto community. Decentralization has always been a core principle for many in crypto.Relying on a single entity for search could concentrate power and data in the hands of a few. Meta’s foray into AI search could set a precedent for other tech giants to follow suit.While innovation is welcome, the shift towards centralization poses challenges for an open internet. The crypto community values transparency and distributed control. As Meta expands its AI capabilities, it’s crucial to advocate for decentralized alternatives.Embracing decentralized search solutions can ensure that power remains with the users. Crypto projects can lead the way in creating open, transparent search technologies. By pushing for decentralization, we can maintain the integrity of the internet ecosystem.Let’s encourage the development of search engines that respect privacy and user autonomy. Together, we can build a future where technology serves the many, not just the few. Stay vigilant and continue supporting decentralized innovations.The battle between centralization and decentralization is shaping the future of tech. Meta’s latest move underscores the importance of advocating for an open, decentralized web. Let’s keep pushing for a future where power is distributed and innovation thrives.

4

One function that has greatly improved in GPT-style AI, is its (now seamless) ability to take in and discuss the content from a URL pasted into a prompt.And a great use-case for that, is instant landing page evaluation for quick tips on improving CRO.As with any prompt like this, improve its effectiveness by adding detail about your target market / ideal customer / nuances about your product or service.🌟 BONUS TIP… and when it comes to finding the perfect content to add to your landing page, one great place to look is in reviews from your own customers. Look through them and see what common themes emerge about what it is that customers *really value* about your offering…So why not paste in the URLs of a few pages of reviews and ask ChatGPT to do that analysis for you… and distil those learnings straight into some finely-tuned benefit text!

6

Introducing Morfless, one of Antler's up-and-coming startup investments using AI to cut cloud costs.As cloud adoption continues to accelerate, so does the complexity and cost associated with managing cloud infrastructure.Morfless was founded by Ian Corbally and Domagoj Filipovic to address this issue head-on.The duo realised the pressing need for a platform that could automate the tedious tasks involved in managing cloud spend, thereby improving operational efficiency and reducing costs.Over the past 12 months, the startup has made significant strides, successfully developing and launching its AI-powered platform, gaining traction among a wide range of Australian and New Zealand sectors – from the largest national media websites to marketing agencies and software companies.Read more in Startup Daily here: https://lnkd.in/gHFjU_SM

50

9 Comments

AI pricing is hard. Customers (esp enterprise) need to do annual budgeting. They won’t sign up for “who knows what it will cost???” But the AI provider (whether OpenAI or something built on it) have highly variable costs based on usage. In addition, if AI increases productivity, it may **reduce the number of user accounts a company buys**. As you provide more value, you (as the AI provider) can see less revenue. No one has entirely solved this.

4

3 Comments

Want to learn how to handle massive traffic spikes and real-time personalization at scale?Check out this new case study on how Meesho leveraged Redis on Google Cloud to keep their shopping platform fast and reliable for over 165 million users.Here's what you'll learn:- How Meesho overcame critical infrastructure challenges during peak sales seasons. ✅- Why they chose Redis to achieve sub-millisecond latency and real-time features. 🔥- How Redis helped them scale to handle 100 million clicks per second during their "Spin the Wheel" game. 🎡- How they reduced monthly operational expenses by 90% with a more cost-effective architecture. 💰Building a tech stack that scales with you is essential for developer success. 💪 This case study is a great example of how Redis on GCP can help you achieve that.Read the full story here: https://lnkd.in/dNujg9WF https://google.smh.re/4d77

3

Monetizing generative AI functionality will be a key challenge in 2025.Ibbaka's second AI Monetization Survey digs deep into how companies are approaching this, probing on market strategy, growth motion, investment and pricing.Please take the survey and contribute your insights.The 2024 survey was taken by 300 companies and has been downloaded more than 1,000 times.https://lnkd.in/gWyGzAet #AI #pricing #monetization

1

Google just announced at their I/O developer conference that they are officially rolling out SGE (Search Generative Experience) today!Understanding Google's SGE:Google's Search Generative Experience is an advanced feature that leverages AI to provide more comprehensive and contextually relevant search results. It integrates elements of generative AI to enhance user interactions with search, offering richer and more detailed responses to queries. This evolution aims to improve user experience by delivering more personalized and immediate information.How this might impact you:SGE has the potential to shift how users interact with search results, emphasizing the importance of high-quality, contextually relevant content. It will likely prioritize content that is informative, engaging, and answers user intent more effectively.At HV, we have been preparing for this and have refined our approach and strategies to ensure our client's businesses continue to thrive in the evolving search landscape. Some high level ways in which we have prepared for this are:1. Content Strategy Optimization: We're focusing even more on in-depth, authoritative content by enhancing long-form pieces, implementing structured data, and optimizing for conversational queries.2. Technical SEO Enhancements: Our thorough audits have additional focus on improved site speed, mobile usability, internal linking, and indexing.3. Data-Driven Insights: Leveraging AI tools help us adapt to search pattern changes, identify content opportunities, and monitor performance in real time.4. Continuous Learning and Adaptation: Our team stays updated with SGE and AI developments through industry webinars and training.5. Client Collaboration: We tailor strategies to meet your specific needs through a collaborative approach.For more details message me and I will be glad to expand.

18

3 Comments

Another interesting perspective from Kyle Poyar on why there is a lack of innovation in the pricing of AI applications.I think Kyle is being gentle here.There is not only a lack of innovation in pricing, there is a lack of innovation in how value is being created and shared.It is still a few months before Ibbaka begins working on its AI monetization in 2025 survey, but when we do this will be a main focus. How is value being created. Without value creation there is not much need for pricing innovation.

3

1 Comment

OpenAI about to launch search 2.0?Are we about to see search getting disrupted AND commoditized? Perplexity attacked weakness in Google’s business model. Positioning on knowdge based search (and not just links) was quite good imho. Now Meta launched native search in WhatsApp. And, now openAI rumoured to be announcing their own search engine (details emerging here https://lnkd.in/gAQZ_Dbm)Funny feeling: in this 4D game, the winner seems to be Bing. Perplexity, openAI all are using them for serp. Ah well we did at least make one bet right for quickads.ai by optimising our PLG for Bing Seo (and not Google). No wonder we get as much as 20% of traffic from them. If anyone really understands deeply Bing Seo then I will love to chat and work together! We are hacker by design and choice :)

19

6 Comments

Seeing strong results lately with two tactics:Product catalogs with cost capAnd product catalogs with a ROAS restriction.When you combine these with cold traffic driven by compelling videos, the results are looking really solid.For one client in the last 7 days, we spent $11,025 and pulled in $99,187 in sales. That’s a 9x ROAS.It’s about finding the right balance between strategy and creative.Focus on scaling smart, not just throwing money at the problem.Have you tried this approach? What’s been your biggest win lately?

227

5 Comments

This Is The Change That Is NeededGoogle's algorithms are literally created to provide a good user experience and is evaluated to that standard. From Google's perspective the algorithm is working if user satisfaction metrics say it's working, wholly indifferent of the publisher's viewpoint as search traffic diminishes with each new Google Search Feature.The disconnect in how Google measures success and the indifference to publishers is built into Google Search. It starts with Google's indifferent mission statement to make information "universally accessible and useful" and ends with the rollout of an algorithm that is 100% indifferent to how publishers experience it. https://lnkd.in/g3FwSGHP

18

2 Comments

Are Generative AI Drawbacks Your Big Differentiation Opportunity?Struggling to find your differentiation in the generative AI space? What if the so-called 'flaws' in AI are actually your secret weapon for differentiation?Let's flip the script on three common AI challenges:1. Prompt Engineering: Embrace the User’s BurdenEver feel like you're playing a guessing game with AI, needing to experiment with the inputs till you get a satisfying response? That's prompt engineering at work. While essential for quality outputs from your LLMs, it can be incredibly frustrating for users.Your move?Develop smart systems that intuitively decode user intent, even from the vaguest prompts. Think multi-prompt strategies and context-aware responses that learn from user history. Suddenly, you're not just another AI tool – you're the one that "gets" your users.2. Hallucination: From Glitch to Unlocking CreativityAI hallucinations – those intriguing yet sometimes annoying fabrications – are usually viewed as liabilities. But what if they could be a hidden wellspring of creativity?For example, imagine you tapped into your model’s hallucination to develop an Ethical Dilemma Simulator for workplace training, or a Strategy Game with "what-if" market scenarios.The key? Clear labeling AI generated content, user controls  for adjusting the level of "creativity" or unexpectedness, and rock-solid ethical guidelines. You're not just sidestepping pitfalls; you're advancing interactive learning by embracing AI hallucinations.3. Small Language Models: Less Can Be MoreIn AI, bigger models isn't always better. Small Language Models (SLMs) with the right pruning and quantization, can provide superior efficiency, privacy, and specialized performance without sacrificing on accuracy.An AI that runs locally on devices offers ironclad data security and lightning-fast responses. For enterprises paranoid about data privacy (aren't we all?), this could be the clincher. Invest in optimizing these compact powerhouses to meet critical user considerations like latency and battery performance, and you might just corner the market with your privacy friendly, "snappy and secure" AI solutions.The bottom line? In the AI gold rush, your differentiator might just be hiding in plain sight – in the very challenges others are scrambling to overcome.I'd love to hear from you: How are you planning to differentiate your product or service with generative AI?

10

1 Comment

Does anyone really use Last Paid Click attribution these days? I’d be shocked if they do. However I found, empirically, that LPC can be a fair attribution for affiliate marketing, specifically for traffic from good old “top 10 ranked “insert product name” LPs. Never researched the reasons for it, my hypothesis is of course the social proof bias. Whilst one should never use LPC when comparing different acquisition channels, it can be useful to compare one affiliate channel to another

The Future of Search and How LLMs will Impact Search BehaviorAs the digital landscape rapidly evolves, the question on everyone's mind is: Are large language models (LLMs) set to disrupt Google Search as we know it?In my latest article for Forbes, I dive into how LLMs like ChatGPT and Google's Gemini are reshaping the way we search for information and engage with content. Could they potentially lead to the downfall of traditional search engines?Here are some key insights from the article:🎯 LLMs as a New Search Paradigm: While search engines have dominated for decades, the advent of AI-powered language models introduces a more conversational and contextual approach to information retrieval.💡 The Shift in Consumer Behavior: Users are increasingly gravitating towards immediate, personalized responses rather than traditional search results.💪 Google’s Response and Future Outlook: Will Google stay ahead of the curve or will it need to adapt to this seismic shift in how users interact with digital content?🔗 Read the full article here: https://lnkd.in/g_7CjpBn As marketers and tech enthusiasts, it’s crucial to stay ahead of these changes. What do you think – is the future of search in the hands of LLMs, or does Google still have a stronghold?#AI #LLM #GoogleSearch #DigitalMarketing #FutureOfSearch #SEO #TechTrends #Innovation

24

Founder Amateur Mode: Randomly executingFounder Beginner Mode: Linearly defining ICP, then working on PMF, then GTMFounder Expert Mode: Working on ICP, PMF, and GTM simultaneously

13

1 Comment

LLM Data Engineering Pattern 3: Aggregating Unstructured Insights at ScaleModern financial analysis often requires synthesizing information from a vast array of unstructured sources—regulatory filings (e.g., 10-K, 10-Q), research reports, market commentary, and even audio transcripts of earnings calls. In many cases, these sources must be combined to form a comprehensive view of an industry’s risk profile. Historically, analysts have relied on manual selection and review of documents, a labor-intensive process that does not scale well in today’s data-rich environment.Consider analyzing credit risk in the U.S. commercial real estate sector:1) Regulatory Filings: An LLM can systematically parse through hundreds of 10-K and 10-Q filings to extract commentary on asset valuations, loan-to-value ratios, or refinancing challenges.2) Research Reports: It can synthesize qualitative assessments from research houses covering property market trends, default probabilities, or vacancy rates.3) Earnings Call Transcripts: It can highlight executives’ forward-looking statements regarding rental income stability, interest rate exposures, or occupancy forecasts.By integrating these unstructured data points, an analyst can address questions like:“Given the commentary on sector vulnerabilities in these filings, research reports, and macroeconomic indicators, what are the key risk factors influencing the commercial real estate market?”Financial people, how are you doing this today?#llm #dataengineering #llmdataengineering #financialdata #finance

20

1 Comment

Is OpenAI using the right product strategy of many little bets? The execution seems a bit scattered. OpenAI Keeps Releasing Prototypes & Previews of Actual ProductsBreaking down o1 and OpenAI's race to get more funding and adapt.OpenAI is at one of its more crucial crossroads as an AI startup maturing into a full fledged company with Billions in revenue.It needs to completely overhaul its corporate structure in order to obtain the funding it requires since it has spent so much in 2024.GPT-4o and o1 do not necessarily compare favorably to the competition and Anthropic’s efforts in particular that are more Enterprise AI aligned. Their use of Chain of Thought in o1 is however fairly fascinating though reviews have been mixed.I have read wildly diverging opinions and hype around it. The benchmark comparisons are no longer an accurate depiction of its potential value. o1 preview is also expensive and clearly targeting an Enterprise audience.OpenAI’s fundamental product strategy is however in question. The cash burn of 2024 raising serious going concerns over the next funding period and corporate overhaul pivot. OpenAI also is trying to raise $5 billion in debt from banks “in the form of a revolving credit facility” according to Bloomberg, and the terms on revolving credit facilities tend to have higher rates of interest.  It needs to raise a huge amount in its next round which might be unsustainable given the rising competition from Anthropic and Google. On the AI chip side ambitions seem even more concerning. The Information also reports that OpenAI is in talks with MGX, a $100 billion investment fund backed by the United Arab Emirates to invest primarily in AI and semiconductor companies, as well as potentially raising from the Abu Dhabi Investment Authority.Over the weekend, Reuters published a report that said any $150 billion valuation would be “contingent” on whether it’s able to rework its entire corporate structure — and, in the process, remove the cap on profits for investors, which are limited to 100x the original stake.I go through its CoT approach to 01 preview and some of my concerns. https://lnkd.in/g2cUUYmn

10

AI memory is fundamental for conversational AI and AI agents. It is critical for building personalized experiences and improving reasoning. Check out Letta’s course to understand how to use AI memory to enhance the customization and quality of your AI app.

15

Good generative AI-era CIOs know the market, the players, the offerings — the actual products, not just the marketing pitch.Bad generative AI-era CIOs listen just to the pitches — and they stop at the hyperscalers and strategy consultants.Good generative AI-era CIOs are hands-on with AI — the APIs, the applications, the studios, the frameworks.Bad generative AI-era CIOs use Copilot to draft their emails — and that's about it.Good generative AI-era CIOs are visionaries who paint a future that all employees — not just in IT — get excited about.Bad generative AI-era CIOs outsource AI vision to Big Tech.Good generative AI-era CIOs build AI teams inside of IT whose job it is to serve the business.Bad generative AI-era CIOs build AI teams who develop AI prototypes untethered to business reality or real users.Good generative AI-era CIOs understand that employee adoption is the gateway to value realization and they partner with their CHROs and business leaders to drive adoption.Bad generative AI-era CIOs don't talk to employees outside of IT about AI.Good generative AI-era CIOs have honest conversations with their CEO and boards about when AI will hit the PnL.Bad generative AI-era CIOs haven't talked to the board about generative AI since 1H'2023.Good generative AI-era CIOs run a cross-functional AI taskforce whose job it is to help the organization responsibly, legally, and ethically develop and use AI applications.Bad generative AI-era CIOs run a cross-functional AI taskforce whose job it is to slow things down until "the market settles."Good generative-AI era CIOs deliver meaningful impact to the organization with generative AI budgets of <$5m.Bad generative-AI era CIOs want to commit tens of millions and issue a press release before POCs are actually in production and getting utilization.Good generative-AI era CIOs understand that scaling and managing AI apps and workflows is going to be just as important as building prototypes, and invest in the tooling and resourcing to do that.Bad generative-AI era CIOs build one company GPT and then lose interest after launching it at the company's AI town hall.Good generative-AI era CIOs know the business needs hands-on AI learnings in order to get their heads around the change management.Bad generative-AI era CIOs think it’s all about their model garden architecture.Good generative-AI era CIOs diversify their approaches to building and deploying AI applications.Bad generative-AI era CIOs just swap LLMs.Good generative-AI era CIOs know that the key to high quality, accurate AI apps and workflows is data — and a lot of that data sits in people's heads and desktops, not just in DBs. Bad generative-AI era CIOs don't want to start on generative AI until the 2-year data transformation project is complete.Good generative-AI era CIOs understand that the speed of the disruption is the disruption.Bad generative-AI era CIOs are still trying to catch their breath.

189

13 Comments

On the A12 WB-Score User Challenge Benchmark leaderboard, the Elo score of 4o-mini is also very high. Its scores in information retrieval, creativity, debugging, mathematics, reasoning, and planning are all in the top tier. Combined with its significant advantages in speed and price, its cost performance is indeed unbeatable.https://lnkd.in/ejRWBB-R

5