- AI Tool Report
- Posts
- 📰 New York Times tests AI ad chatbot
📰 New York Times tests AI ad chatbot
🎖️ AI for Professionals by AI Tool Report
Martin Crowley & Liam Lawson
February 22, 2024
🕵️♂️ Job Board // Pro // Podcast // Tools Database // Accreditations🥇
Thursday’s top story: The New York Times (despite suing OpenAI) is building an AI ad chatbot that will allow advertisers to plan and optimize ad campaigns in real time.
A quick reminder: We’ll launch AI for professionals (the only AI accreditation you’ll ever need) on March 5th, so sign up now, secure your place, and get a 10% discount for early access.
P.S. Click to share this newsletter with a friend 🧑🤝🧑
🌤️ This Morning on AI Tool Report
📰 New York Times tests AI ad chatbot
©️ Google is Reddit’s AI content partner
🚀 Google launches open-source AI models
⏱️ Sora takes 1hr to create 1min of video
💘 Match signs contract with OpenAI
Read Time: 5 minutes
FINANCE
CHATBOTS
Our Report: The New York Times (NYT) is recruiting partners to test a generative AI chatbot (in closed beta) that will recommend the best placement for ad campaigns, making it easier for advertisers to plan and optimize ad campaigns in real-time.
Key Points:
The AI chatbot will provide advertisers with recommendations for where and when an ad campaign could perform best, based on its messaging and campaign objectives.
Advertisers will also use the new tool to target previously unidentifiable niche audiences based on their interests, goals, aspirations, and opinions.
The NYT is experimenting with multiple large language models (LLMs) to power the tool and plans to make it available to marketers in the second half of the year.
Although costs remain unknown, once launched, brands will only have access to the tool via the NYT ad team, who will apply it to ad campaigns.
Why you should care: This development follows the hiring of an editorial director of AI initiatives and reflects the publishing industry's response to the movement away from third-party tracking cookies, by improving real-time ad targeting.
In partnership with Coefficient.
Those tedious exports and imports are… unfulfilling.
That’s why 300K+ Google Sheets and Excel users rely on Coefficient to power their spreadsheets with live data.
Coefficient connects your spreadsheet to any data source in seconds. Import the data you need, and set it on a refresh schedule to power live reports and processes.
Connect & keep your spreadsheet in sync with your CRM, BI tool, database, marketing platforms, accounting systems, and more
Deliver insights into Slack or email
Blend data from multiple sources
Sync data back to your source system from your spreadsheet
Use GPT Copilot to generate formulas, SQL, charts, pivots, and more
The extension costs nothing to use.
Guidde is the secret presentation tool that will 10x your team’s productivity 🤫💡
Adobe AI Assistant will improve digital document interaction
Emails Copilot will enhance email creation processes using AI
FastSDXL uses AI to build ready-to-use model APIs
Cliptics transforms text into lifelike speech
CONTENT DATA
Last week Reddit announced it was negotiating a content deal with a tech giant, and it’s been revealed that Google will sign the $60M deal and use its content to train AI models.
This comes as Reddit seeks to generate new revenue streams in the face of competition for advertising dollars from platforms like TikTok and Facebook.
It also comes amid reports that it's preparing to make its initial public offering filing this week, where it will look to sell around 10% of its shares.
If successful, the deal with Google will prove the growing trend of AI model makers securing deals with content owners to expand the diversity of their training data.
✅ Blog Creation
copy & paste ⬇️
“I'm looking for a [type of blog post] that will clearly explain the features and benefits of my [product/service] to [ideal customer persona] and persuade them to make a purchase with a strong call-to-action."
OPEN-SOURCE AI
Following the release of Gemini, Google has launched two open-source AI models (Gemma 2B and Gemma 7B) that have been built using the Gemini infrastructure.
Both models come with pre-trained and instruction-tuned variants and have been optimized to run on developer laptops and desktops, using NVIDIA chips to amplify performance.
Google claims that Gemma is faster, can run locally on devices, and doesn’t need to share its results with third parties, making it better than other models (like Meta’s Llama 2).
It filters out sensitive data from its training sets and comes with a ‘Responsible Generative AI Toolkit’ that provides a debugging tool, safety classification, and guidance.
TEXT-TO-VIDEO
Red team testers have reported that OpenAI’s new text-to-video AI model, Sora, takes over an hour to produce a one-minute video due to overly long rendering times.
This is due to the amount of data and computations needed—particularly when iterating the video using multiple different prompts—to produce a high-quality video.
However, some testers believe that when compared to the current amount of manual labor needed to produce high-quality videos, 90 hours for a 90-minute film is quite reasonable.
This highlights that, while the potential for AI-generated video creation is evident, addressing efficiency and scalability remains crucial for wider adoption by the public.
Do you think Sora is the reason Sam Altman needs $7 trillion to build his AI chip factories? |
ONLINE DATING
In a ChatGPT-penned press release, dating app Match announced an agreement with OpenAI that will see it receive 1,000 ChatGPT enterprise licenses for its employees.
Employees will use ChatGPT-4 to complete work-related tasks, such as coding, design, analysis, template building, and communication, making teams more productive.
To keep its corporate data protected, only trained and licensed employees will have access to OpenAI’s tools, and they’ll be guided by existing AI principles.
They’re also planning to invest $30M to build AI-powered apps to help with dating profile creation, matching abilities, and post-match guidance across its other dating apps.
In Partnership with Graft.
Product managers, if you spend most of your time searching for answers, updating docs, summarizing customer feedback, and wrangling messy spreadsheets, it’s time for a change.
Let your AI teammate do the work for you:
Effortlessly categorize and analyze user feedback, dramatically reducing your time trying to understand customer needs and sentiments.
Generate and maintain up-to-date product documentation and reports effortlessly, keeping your team informed and aligned.
Instantly find critical data or documents, eliminating the endless search and boosting your decision-making speed.
Organize and prioritize product features and updates seamlessly, ensuring your roadmap directly reflects user demands.
The best part? Your AI Teammate goes beyond routine tasks; it adapts and learns from your specific knowledge and expertise. This trainable AI teammate is designed to complete work as you would, aligning closely with your preferred methods and goals.
We read your emails, comments, and poll replies daily.
Hit reply and let us know what you want more of!
What did you think of this edition? |
