Possibilities:
1) Link building
2) Someone forgot to cancel their recurring SaaS subscription and now it runs on autopilot
3) Agency hasn't updated their processes since 2010
Translation: your insistence on how quality search term matches limits our next quarter profits.
3) Over-relying on AI might just erode our memory, analytical skills, and overall cognitive flexibility.
What do you think? Have you noticed this on yourself? Have you just copy-pasted answers from ChatGPT without thinking about them much?
1) When we offload our thinking to AI, we miss out on deep, reflective thought. It's convenient, sure, but at what cost?
2) Younger people are especially at risk, showing higher AI dependence and lower critical thinking scores.
So, what does this mean for us? It's like our minds are taking a backseat while AI does the heavy lifting. 🚗💨
A few not so promising findings from the study.
The abstract states, "The findings revealed a significant negative correlation between frequent AI tool usage and critical thinking abilities, mediated by increased cognitive offloading."
🧠 AI does to your brain what sitting does to your body
New study "AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking" by Michael Gerlich has alarming results.
www.mdpi.com/2075-4698/15...
Exactly, the relevant conversions your prospects have happen largely outside of your website. As a die-hard and old school SEO guy, I've had to drag myself to be more active on social.
AI so smart it has transcended language
Water as a service, WaaS!
Look forward to seeing my "link articles" in Hub pages getting cited in AI Overviews 🤣
Can't imagine the ads Google shows you based on your "search interest" with all of the examples you've shared recently.
Absolutely, the hack of the week by a guest contributor "definitely-a-bot"
This also happened. Volkswagen tracked location, start and stop times and other data for their cars. They also left the data unprotected and accessible online. Great job!
www.darkreading.com/cyberattacks...
Agree that fake reviews are illegal. I was commenting more from trust point of view. AI written fake reviews are probably one of the fastest ways to lose customer trust - and their business.
There are some things you can measure, and there are some things you can't measure. Paying too much attention to the former you easily destroy the latter. And it's usually the latter that leads to conversions and revenue.
I'm privy to secret, insider knowledge!
My additions to this:
Script_1_new.py
Script_1_new_v2.py
Script_1_final.py
Script_1_final_new.py
This has always baffled me. Why would G rank gated content given that it will be a bad experience for over 95% of their users. In shopping ads they sneakily go and have bots add items into cart for price verification. So they could check for public access if they wanted to.
☝️ is the thing that separates inexperienced SEOs from more senior and "enterprise SEO" folks. Most people can give decent recommendations but understanding what can or should be implemented is harder.
It's been 3 years since I stopped hearing baby shark in my mind - and now you started it again! 😅
In the future it's cheaper for AI to hire humans!
So, should we be dusting off our résumés? Perhaps not just yet. What do you think? Are you ready to welcome our robot overlords, or will you hold on to your swivel chair a bit longer? Let me know! 😉
Oh, and browsing the web? A pop-up asking to download a mobile app had them stumped. Maybe those EU cookie banners are our last line of defense after all! 🍪
Turns out, simple tasks that humans find boring (like admin stuff) are the hardest for AI agents.
And my favorite: an agent was tasked with asking colleague Alex who to introduce themselves to next. Alex provided a name, and instead of reaching out to that person, the agent... checked the task off as complete. Networking level: AI. 🤖
Another agent was supposed to write to `/workspace/answer.docx` but didn't realize it was a Word file (since it wasn't explicitly told). It treated it as plain text instead. Instructions not clear enough? 🤷♂️
Now, where did our digital coworkers stumble?
In one instance, an agent couldn't find the right person on RocketChat, so it just renamed another user to match. Talk about taking initiative!
And it wasn't cheap! Claude Sonnet averaged 29 steps per task at $6.34 a pop. GPT-4o, perhaps wisely, gave up sooner, costing $1.29 per task.
The results? Well, let's just say humans aren't obsolete yet.
The star performer, Claude Sonnet 3.5, completed a whopping 24% of tasks, scoring 34% with partial completions.
GPT-4o managed 9% completion with a 17% score.
