Welcome back, AI Enthusiasts!
In today’s AI summary rundown:
Historic Revenue Milestone
Alphabet’s total annual revenue surpassed $400 billion in 2025 for the first time, reflecting strong performance across core businesses.Q4 Growth Numbers
In the fourth quarter, revenue reached roughly $113.8 billion, up about 18 percent year-over-year, with profit also rising significantly.YouTube and Search Drive Earnings
YouTube’s combined advertising and subscription revenue topped $60 billion annually, while search and ads remained core revenue engines.Cloud and AI Adoption
Google Cloud revenue climbed around 48 percent, with strong enterprise demand tied to AI infrastructure and solutions.AI Investment Expansion
Alphabet plans to roughly double its AI spending in 2026, allocating as much as $175–$185 billion to capital expenditures to support AI growth and infrastructure.
Read time is 4 min..
AI Insights
OpenAI and Anthropic Take AI Chatbots To The Super Bowl Advertising Stage
Super Bowl Ad Campaign Launch
Both Anthropic and OpenAI entered the high-profile Super Bowl advertising arena, drawing tech and mainstream attention.Anthropic’s Message
Anthropic’s spots highlight a contrast: Claude remains ad-free, while other AI tools (implicitly ChatGPT) are depicted as introducing intrusive ads into conversations.Ad Content and Reach
The campaign includes multiple 30-second commercials aired during and around Super Bowl 60, using humor to showcase the idea “Ads are coming to AI. But not to Claude.”OpenAI CEO Response
OpenAI’s Sam Altman called the ads funny but “dishonest,” defending ChatGPT’s ad strategy and saying it would avoid disruptive mid-conversation placements.Advertising Debate in AI
The ads reflect a broader industry debate over monetization versus user experience, as AI developers balance revenue models with trust, privacy, and engagement concerns.
AI Training: AI Tutorial of the Day
Explain Long-Term Thinking as Delayed Reward Optimization
One experiment I’ve been exploring on thareja.ai is using AI concepts to explain human behavior.
Reframing long-term thinking as delayed reward optimization helps make patience, discipline, and planning easier to understand through a reinforcement-learning lens.
Start a New Chat
Open thareja.ai and begin a fresh conversation.
A clean slate helps the model focus on long-term incentives rather than short-term context.
Switch Your AI Model
Click the ( + ) icon next to Automatic.
Choose Claude for this experiment.
Why Claude?
Claude excels at:
Structured reasoning
Explaining trade-offs clearly
Maintaining logical progression over time
Perfect for modeling delayed rewards.
Try This Prompt & Observe the Output
Prompt used:
“Explain long-term thinking as delayed reward optimization in reinforcement learning. Compare short-term gratification to immediate rewards and long-term goals to delayed rewards, using simple real-life examples.”
Model Used:
Claude
AI Response (Excerpt):
“Immediate gratification is like choosing a small reward now, while long-term thinking prioritizes larger rewards later. In reinforcement learning, agents that optimize for delayed rewards outperform those chasing instant gains, even if progress feels slower at first.”
Why This Experiment Works
Makes patience and planning easier to visualize
Connects human behavior to AI training dynamics
Clarifies why short-term wins can sabotage long-term outcomes
Each model brings a different mindset. GPT favors storytelling and metaphor. Claude focuses on clarity and structure. Testing both on thareja.ai helps you find what actually works.
Happy promoting!
Exclusive Member Deal - 20% OFF
Superpower AI Bundle Access 50+ Major LLMs
One Subscription. $20/mo $16/mo with code SUPERPOWER20
AI-Generated Image of the Day
Prompt: A bustling painting class in a well-lit studio. The room is filled with students of diverse ages and backgrounds, each immersed in their creative process. Some are working on easels with canvases, while others are experimenting with different painting techniques. The instructor, a seasoned artist, moves around the room, offering tips and encouragement. The walls are adorned with completed artworks, showcasing a variety of styles from abstract to realistic.
Tip: the more specific the better
Jump between models on thareja.ai without losing speed. Nano Banana helps you create eye-catching visuals that are made to perform on social platforms.
Meme of the Day
Question of the Day
What is Deep Learning a subset of?
That’s it for today’s news in the world of AI!
If you have anything interesting to share, please reach out to us by sending us a DM on Twitter: @dthareja or email me at [email protected]. How was today's newsletter?
Feedback helps us improve!
Thanks for reading. Until next time!
p.s. if you want to sign up for this newsletter or share it with a friend or colleague, you can find us here.