The AI news is shifting – we’ve moved from rapid-fire product announcements to predictions and regulation.
Let’s hit that regulation area first. India’s Ministry of Electronics and IT said last week that they do not plan to regulate the growth of AI within the South Asian market, identifying the sector as a “significant and strategic” area for the nation. The ministry asserted that the expansion of AI would have a “kinetic effect” on entrepreneurship and business development in India.
Meanwhile, the Washington Post dives into Congress’s willingness to investigate AI regulation. The thesis is that lawmakers, burned by not regulating social media prior, are looking to not fall into the same trap.
The Center for AI and Digital Policy (CAIDP) asked the Federal Trade Commission to investigate OpenAI for violating consumer protection rules, arguing that the organization’s rollout of AI text generation tools has been “biased, deceptive, and a risk to public safety.”
On the technology front, there’s another debate – open-source large language models versus their closed-source competitors. VentureBeat dives into options like Dolly, based on Alpaca, another open-source LLM, or Meta’s offerings in the open-source world with LlaMA by making their weights available for researchers and academics. Or names like Vicuna and Koala/
Because the adoption continues to drive – a new survey data from WordFinder finds that when asked how often they use ChatGPT, 46% said they’d used it only once or twice. Out of the rest, 10% use it several times a month, 20% use it once a week, 19% use it several times a week, and 5% use it daily. More than half of the people in each age group use ChatGPT regularly (at least several times a month), with baby boomers at 54%, Gen X at 51%, Millennials at 55%, and Gen Z at 61%.
Those in the tech field are the biggest users of ChatGPT, with 62% of them using it regularly. Some 59% of healthcare workers and 56% of those in marketing also use it regularly. Less frequent users included people in construction at 51%, those in business services at 50%, and those in education at 50%.
If you want to explore some of the impacts, I’ve included a link to a Vice article about Reddit mods fighting ChatGPT content. Shocking no one, it’s already a significant problem.
Why do we care?
I’m somewhat skeptical of Congress’ ability to act here, although I wish they would. Privacy laws would go a very long way. Now, I noted that the AI group filed because of the who – the FTC. There’s an existing framework there that will likely be applicable, and while it’s to be tested in the courts, look there for guidance.
Those adoption numbers continue to impress. More than anything, that’s why to note what’s happening.