New rating ElevenLabs
ElevenLabs
D
Added to the Index
Your voice becomes a model in somebody else's account, and you cannot take it with you when you leave.
What changed
AI products change fast, so the ratings do too. Every revision is logged here with the date, the direction the rating moved, and the reason. Nothing gets quietly rewritten.
16 ratings tracked 7 revisions
Latest
October 2026
New rating Apple
Apple Intelligence
B
Added to the Index
The only AI on this Index whose privacy claims you can go and check yourself. The accuracy claims are another matter.
September 2026
New rating Flock Safety
Flock Safety
D
Added to the Index
Twenty billion vehicle scans a month, searchable by 5,000 police agencies. There is no opt-out, because you were never asked.
Updated Adobe
Adobe Firefly
Grade C → D
risk 5.8 → 6.6(Worse)
Adobe took Firefly's Generate Music, Generate Speech, and Generate Sound Effects out of beta on August 20, 2026, routing speech through Adobe's own model or a named third party, ElevenLabs, and made the multi-step Firefly AI Assistant free to use with a daily generation allowance. That extends displacement risk from illustrators and stock photographers into voiceover artists, composers, and sound designers, and hands more of the actual editing to an agent working with less step-by-step oversight. Job threat moved from 8 to 9, autonomy from 4 to 6, and privacy from 4 to 5 to reflect audio and voice-sample data now reaching a named third-party processor.
New rating Alphabet
Waymo
C
Added to the Index
Measurably safer than the drivers it replaces, wrapped in a 360-degree camera that never stops recording the street.
Updated Perplexity AI
Perplexity
Grade C → D
risk 5.4 → 6.2(Worse)
Perplexity spent 2026 building Computer, an agent with standing access to a user's files and apps: a Mac version shipped in April, a $20-a-month Pro tier opened in May, and a local, on-device version called Portable Computer reached Windows RTX PCs on September 14. The company now credits that product line with most of its revenue growth, and its CEO has said the goal is to run businesses as autonomously as possible while dismissing the resulting job losses as an acceptable trade. Autonomy moved from 6 to 8, job threat from 6 to 7, and lock-in from 3 to 4, taking the grade from C to D.
New rating Midjourney, Inc.
Midjourney
D
Added to the Index
Your prompts and pictures are public unless you pay $60 a month, and three Hollywood studios are asking a court to switch it off.
Updated Tesla
Tesla Autopilot / FSD
Grade F held
risk 7.8 → 8.2(Worse)
Tesla began pulling the human safety monitor out of some Austin robotaxis in January 2026, so a subset of paid rides now run with nobody aboard able to intervene, and NHTSA escalated its FSD visibility probe to an Engineering Analysis that March, covering 3.2 million vehicles and the stage that usually precedes a recall. A Florida jury's $243 million Autopilot verdict survived Tesla's appeal in February, and Tesla has since settled at least four more wrongful-death Autopilot and FSD suits rather than face another jury. Autonomy moves from 9 to 10, job threat from 7 to 8.
New rating Otter.ai, Inc.
Otter.ai
D
Added to the Index
The bot your colleague invited to the call, recording everyone else on a consent nobody but the host gave.
Updated Google
Google Gemini
Grade D held
risk 7.2 → 7.4(Worse)
In May 2026 Google folded its experimental Project Mariner browsing agent into Gemini and rolled out Gemini Agent automation on Android, letting the assistant fill shopping carts, check ticket availability and complete purchases on its own. An April 2026 update to the Agent Payments Protocol added a Human Not Present mode that lets Gemini buy something the moment it becomes available without a person confirming the purchase. Autonomy moved from 7 to 8.
New rating Microsoft
Microsoft Recall
B
Added to the Index
A screenshot of your screen every five seconds, indexed and searchable. It never leaves the laptop, which is the only reason this is not an F.
August 2026
Updated Microsoft
GitHub Copilot
Grade C held
risk 5.2 → 5.8(Worse)
Since April 2026 GitHub trains on prompts and code by default on the Free, Pro and Pro+ plans, which had required opt-in before. Entry-level developer postings were down 67 percent from 2022 levels, and new agents now run on a schedule and check their own fixes without a developer in the loop. Privacy 5 to 6, job threat 7 to 8, autonomy 5 to 6.
New rating Meta / EssilorLuxottica
Ray-Ban Meta Smart Glasses
C
Added to the Index
A camera and microphone worn on your face, wired to Meta. The people around you never agreed to any of it.
Updated OpenAI
ChatGPT
Grade D held
risk 6.6 → 7.4(Worse)
OpenAI turned on ad tracking by default for free users, a class action over query data shared with Meta and Google survived a motion to dismiss, and Agent Mode shipped autonomous multi-step tasks with memory that persists between sessions. Privacy 7 to 8, autonomy 6 to 8, lock-in 7 to 8.
New rating Workday
Workday AI Recruiting
D
Added to the Index
The screening layer between you and a billion job applications, with its bias testing sealed under attorney-client privilege.
Updated Amazon
Amazon Alexa
Grade D held
risk 6.2 → 6.6(Worse)
Alexa+ reached every US user in February 2026 and has since spread to Australia, a browser version and car dashboards. It keeps memory across sessions and completes tasks on its own rather than answering one command at a time, and Amazon retired on-device voice processing in March 2025, so requests now leave the house. Autonomy 7 to 8, lock-in 8 to 9. The grade holds at D.
New rating Adobe
Adobe Firefly
D
Added to the Index
Trained on data it actually paid for, which is rarer than it should be. Getting back out of Adobe is the expensive part.
New rating Perplexity AI
Perplexity
D
Added to the Index
Now ships an agent that works your files and apps around the clock, while still being sued by nearly every major newsroom.
March 2026
New rating Amazon
Amazon Alexa
D
Added to the Index
A corporate spy you bought with your own money, placed in your bedroom, and thanked for the weather report.
New rating OpenAI
ChatGPT
D
Added to the Index
Knows more about you than your therapist. Reports to a company that wants to be God.
New rating Microsoft
GitHub Copilot
C
Added to the Index
Your AI pair programmer that learned to code by reading everyone else's code without asking.
New rating Google
Google Gemini
D
Added to the Index
Google already had all your data. Now it has a chatbot that can actually use it.
New rating Tesla
Tesla Autopilot / FSD
F
Added to the Index
A $15,000 promise of self-driving that's killed people and still calls itself 'Full Self-Driving.'