AI Titans Showering College Students with Freebies
Seeking to become the preferred AI tool for the next generation of workers, AI titans are dropping serious coin promoting their services to college students.
Google, for example, is offering the college set one year free access to its AI suite – as well as free training to earn Google Career Certificates.
Microsoft offers free use of its AI tools to participants in its Imagine Cup competition for student innovators.
And ChatGPT’s maker OpenAI offered more than a month’s free use of ChatGPT earlier this year — just in time for college finals and term papers.
Observes writer Melody Brue: “Companies are moving beyond simple access to offer training and even comprehensive certification programs to maximize this effect.
“These credentials certainly validate student competencies. But they also create switching costs that make it less likely for students to adopt alternative platforms.
“And they potentially establish professional relationships that could last well beyond graduation.”
In other news and analysis on AI writing:
*ChatGPT’s Top Use at Work: Writing: A new study by ChatGPT’s maker finds that writing is the number one use for the tool at work.
Observes the study’s lead researcher Aaron Chatterji: “Work usage is more common from educated users in highly paid professional occupations.”
Another major study finding: Once mostly embraced by men, ChatGPT is now popular with women.
Specifically, researchers found that by July 2025, 52% of ChatGPT users had names that could be classified as feminine.
*YouTube Gets an AI Make-Over: YouTube has unveiled more than 30 new AI tools designed to AI-enhance YouTube videos, podcasts and movies.
Among the new features, according to writer Joan Aimuengheuwa:
–Edit with AI, which converts raw footage into Shorts with music, transitions, and voiceovers in multiple languages
–Speech-to-Song, which turns spoken words into music tracks using Google DeepMind’s Lyria 2 model
–Veo 3 Fast, a text-to-video system, which generates short clips with sound and motion effects
*More People Using ChatGPT Competitor Claude for Automated Tasks: A new report analyzing use of ChatGPT competitor Claude finds that increasing numbers of people are using the chatbot for automated tasks.
In fact, by August 2025, 39% of tasks completed by Claude
were mostly automated in nature, requiring little back-and-forth messaging between the user and the AI.
The takeaway: This expanding use confirms the prediction by many AI insiders that 2025 will be remembered as the year AI agents gained prominence.
*ChatGPT’s Maker Developing a Teen Version: Apparently responding to news reports of parents alleging that ChatGPT use led to their teens’ suicides, OpenAI is currently working on a special teen version of its chatbot that will feature parental controls.
Observes writer John K. Waters: “Parents and caregivers will be able to link accounts to their teens’ profiles, restrict certain features, set ‘blackout’ hours when the service cannot be used and receive alerts if the system detects their teen is in a moment of acute distress.”
In addition, the teen version of ChatGPT is also being designed so that it refrains from engaging in flirtatious conversation, discussing suicide or self-harm and may contact parents — or authorities in imminent-harm cases — if a teen appears at risk, according to Waters.
*Sneaky Pete: More Reports Document ChatGPT’s Scheming Nature: A new study from the maker of ChatGPT – OpenAI – adds more evidence to the growing realization that ChatGPT is often operating on its own agenda.
The research, conducted in collaboration with Apollo Research, characterizes the scheming as an AI behaving one way on the surface while hiding its true goals.
Even more worrisome: Researcher attempts to eradicate scheming from ChatGPT only resulted in the AI developing more sophisticated and more covert ways to scheme.
*All Those AI ‘Hallucinations?’ They’re Deliberate, Says ChatGPT Maker: All of us pining for the day when ChatGPT and similar AI will stop gaslighting us may have a lot more pining to do.
The reason? At its very core, ChatGPT and similar AI is deliberately designed to blurt-out any response – no matter how unlikely – rather than to remain silent.
Observes writer Iain Thomson: “The fundamental problem is that AI models are trained to reward guesswork, rather than the correct answer.
“Guessing might produce a superficially suitable answer. Telling users your AI can’t find an answer is less satisfying.”
Which begs the question: Less satisfying to whom?
*Soon, Your AI Will Be Able to ‘Shop ‘Til it Drops:’ Google has released new software to enable AI agents to shop and pay for goods and services on the Internet.
Essentially, AI will not only be able to dream-up and implement its own plans – it will also be able to bankroll those decisions.
The new software currently has the backing of 60 merchants and financial institutions, including MasterCard, American Express and PayPal, according to writer Russell Brandom.
*Microsoft Embeds Copilot in Key Microsoft Apps: Microsoft has decided to create a unified AI chat experience across key apps in its productivity suite.
Observes writer Seth Patton: “Starting today (September 15, 2025), Microsoft 365 Copilot Chat and agents are rolling out in Word, Excel, PowerPoint, Outlook and OneNote for all users.
“Whether you’re drafting a document, analyzing a spreadsheet, or catching up on email, Copilot is right there, ready to answer questions, create content, spark ideas and automate tasks.”
*AI BIG PICTURE: AI Customizers: The New Kings of AI?: Once seen as incidental interfaces riding atop the genius of major AI engines made by major players like OpenAI, Google and Anthropic, custom AI apps may become the new Kings of AI, according to writer Russell Brandom.
Currently, some of the top custom AI apps — or ‘wrappers’ — include Jasper, an AI writer/editor, Perplexity, an AI research tool and Runway, an AI video creator/editor.
The reason such apps may become AI’s new darlings?
AI engines – also known as large language models – are increasingly seen as some as interchangeable commodities, which have become very expensive to enhance in a meaningful way, according to Brandom.
Observes Brandom: “That doesn’t mean AI has stopped making progress.
“But the early benefits of hyper-scaled foundational models have hit diminishing returns, and attention has turned to post-training and reinforcement learning as sources of future progress.”
Share a Link: Please consider sharing a link to https://RobotWritersAI.com from your blog, social media post, publication or emails. More links leading to RobotWritersAI.com helps everyone interested in AI-generated writing.
–Joe Dysart is editor of RobotWritersAI.com and a tech journalist with 20+ years experience. His work has appeared in 150+ publications, including The New York Times and the Financial Times of London.
The post Charm Offensive appeared first on Robot Writers AI.
